diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 41c5dc68fa72c68457e24ffa7e422780677905b1..a0c438ddecbcb44ffa4995454cb2441c7e6c30ce 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": "20260513T180036Z", - "generated_at": "2026-05-13T18:06:49Z", + "snapshot_id": "20260514T000044Z", + "generated_at": "2026-05-14T00:09:18Z", "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.", @@ -1620,6 +1422,76 @@ } ] }, + { + "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": 43824, + "canonical_pr_number": 43836, + "issue_numbers": [ + 43824 + ], + "pr_numbers": [ + 43836, + 43842 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "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.", @@ -2117,16 +1989,6 @@ "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, @@ -2136,6 +1998,15 @@ "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, shared_issue_target evidence." + }, { "cluster_id": "cluster-43979-11", "canonical_pr_number": 44007, diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index 85ca535b4f8c6269b33261aa6bf802cfaa768718..a93285234a17039820f3ee4c97b484e0a6013294 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": "20260513T180036Z", - "generated_at": "2026-05-13T18:06:49Z", + "snapshot_id": "20260514T000044Z", + "generated_at": "2026-05-14T00:09:18Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -131,406 +131,310 @@ "issue:41093|issue:43638", "issue:43792|issue:45901" ], - "prepared_review_unit_hash": "24c997d316311c5bd1cf4780c95aa95609e8fe53a2279f94318216256338470c", + "prepared_review_unit_hash": "49ac566543f39b872fa7fc45b0b8d1e5a252eca12eec40fe37b4bec356f92c44", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 62, - "serialized_chars": 46365, - "estimated_input_tokens": 11592, - "estimated_eval_tokens": 23440 + "serialized_chars": 46370, + "estimated_input_tokens": 11593, + "estimated_eval_tokens": 23442 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a collection of distinct bug reports, with only a few true duplicates: the tokenizer regression reports, the Qwen3.5 MoE rotary-embedding/offload reports, and the DeepSpeed ZeRO-3 BERT loading reports.", - "confidence": 0.81, - "canonical_issue_reason": "issue:45920 is the broadest representative of the tokenizer regression cluster: it describes wrong token IDs across multiple model families, not just Granite, and it is still open with additional references.", + "summary": "This is a heterogeneous cluster; most pairs are only superficially similar. The strongest duplicate groups are the ZeRO-3 Bert load failures, the Qwen3.5Moe rotary-embedding device/materialization reports, and the v5 AutoTokenizer token-ID regression reports.", + "confidence": 0.69, + "canonical_issue_reason": "issue:45920 is the broadest, most representative current report of the v5 AutoTokenizer token-ID regression across multiple model families; it subsumes the narrower Granite-specific report in issue:45812.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best global issue candidate because it captures the widest version-5 tokenizer regression and is phrased as the most general, reusable tracking report.", + "best_issue_reason": "issue:45920 is the best canonical issue because it describes the widest impact and the clearest reusable fix target among the accepted duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different failures: a missing Qwen3.5 config field vs a chat/completions streaming inference error." + "reason": "Different models and different failures; not the same bug." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Both involve GPT-OSS/MoE-ish model issues, but the failures are unrelated: AMD GPU runtime problems vs an MPS histogram op limitation." + "reason": "Unrelated hardware/backend errors with no shared code-path evidence." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Mistral-3 dtype mismatch and GPT-OSS on AMD GPUs are different bugs and code paths." + "reason": "Different model families and different failure modes." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Autocast dtype mismatch is a different failure from GPT-OSS GPU compatibility issues." + "reason": "Both are dtype/device-style errors, but the model paths and bugs differ." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model families and different runtime errors; not the same underlying bug." + "reason": "Different model-specific failures; too broad to merge." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime failure is unrelated to GPT-OSS AMD GPU support." + "reason": "Unrelated model/runtime issues." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "DeepSpeed ZeRO-3 collective mismatch and ONNX export failure are unrelated." + "reason": "DeepSpeed collective mismatch vs ONNX export are unrelated." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "ModernBERT ONNX export and Qwen3-MoE FP8 failure are unrelated model-specific issues." + "reason": "Different subsystems and different concrete failures." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "ONNX export vs auto device map cuda assert are different problems." + "reason": "ONNX export and auto device mapping are unrelated." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are v5 docs cleanup items, but they target different removed references and are not clearly the same concrete fix." + "reason": "Both are docs-related, but they target different removed v5 features." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ONNX export failure and GenerationMixin import regression are unrelated." + "reason": "ONNX export vs import error; not the same issue." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Mask-shape index error and Gemma4 PLE device mismatch are different failures." + "reason": "Different models and different runtime paths." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both are index/mask errors, but the error shapes and contexts differ; not enough to treat as one bug." + "reason": "Same generic IndexError shape pattern, but not enough evidence of the same root cause." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Different runtime errors in similar areas, but not the same underlying issue." + "reason": "Same generic IndexError pattern, but distinct repros and model paths." }, { "left": "issue:44805", "right": "issue:45161", "accept": false, - "reason": "Mask-shape mismatch and GPT-OSS tensor-parallel failure are unrelated." - }, - { - "left": "issue:44805", - "right": "issue:44918", - "accept": false, - "reason": "Different shape/runtime failures in different trainer/model paths." - }, - { - "left": "issue:34634", - "right": "issue:35545", - "accept": false, - "reason": "BarkProcessor voice_preset and ModernBERT ONNX export are unrelated." - }, - { - "left": "issue:30277", - "right": "issue:43638", - "accept": false, - "reason": "Both touch training/loading infrastructure, but the concrete failures and model paths differ." - }, - { - "left": "issue:43596", - "right": "issue:43638", - "accept": true, - "reason": "Same DeepSpeed ZeRO-3 index-out-of-bounds failure when loading BertModel variants." + "reason": "Different models and unrelated failure classes." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger cross-entropy behavior and GPT-OSS tensor-parallel failure are different bugs." + "reason": "Different feature areas; no concrete shared bug." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "Incorrect detokenization and LoRA merge/save result mismatch are different problems." + "reason": "Tokenizer detokenization bug vs LoRA merge/save mismatch are not the same defect." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Qwen NVFP4 runtime error and LoRA save/reload mismatch are unrelated." + "reason": "Different model and different regression." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "FSDP activation-checkpointing recomputation mismatch is a different issue from the mask-shape IndexError." + "reason": "Both are shape-related, but they describe different training/runtime problems." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Both involve DeepSpeed, but they are different concrete failures on different models and code paths." + "reason": "Different DeepSpeed errors and different affected code paths." }, { "left": "issue:45859", "right": "issue:45902", "accept": true, - "reason": "Same Qwen3.5 MoE rotary-embedding bug around device/offload materialization and missing initialization." + "reason": "Same Qwen3_5MoeTextRotaryEmbedding `inv_freq`/device-movement bug, surfaced via CPU offload vs `meta -> to_empty(cuda)` materialization." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "ZeRO-3 BERT loading and GPT-OSS tensor-parallel issues are unrelated." - }, - { - "left": "issue:34928", - "right": "issue:36331", - "accept": false, - "reason": "Activation-checkpointing recomputation mismatch and compute_loss signature regression are unrelated." - }, - { - "left": "issue:43596", - "right": "issue:45161", - "accept": false, - "reason": "BERT ZeRO-3 loading failure is unrelated to GPT-OSS tensor-parallel behavior." - }, - { - "left": "issue:43716", - "right": "issue:45507", - "accept": false, - "reason": "Mistral-3 dtype mismatch and GraniteMoEHybrid invalid-method call are different bugs." - }, - { - "left": "issue:39290", - "right": "issue:45081", - "accept": false, - "reason": "Gemma3/vLLM sliding_window config error and Mistral tokenizer regex patch crash are unrelated." - }, - { - "left": "issue:45823", - "right": "issue:45874", - "accept": false, - "reason": "Gemma4 device mismatch and expensive reverse embedding lookup are different forward-path issues." - }, - { - "left": "issue:43425", - "right": "issue:44292", - "accept": false, - "reason": "Torch compatibility and Qwen NVFP4 runtime failure are unrelated." - }, - { - "left": "issue:43828", - "right": "issue:45161", - "accept": false, - "reason": "Autocast dtype mismatch is not the same as GPT-OSS tensor-parallel failure." - }, - { - "left": "issue:34928", - "right": "issue:41720", - "accept": false, - "reason": "FSDP checkpointing mismatch and Qwen3 auto-device-map CUDA assert are different bugs." - }, - { - "left": "issue:44918", - "right": "issue:45161", - "accept": false, - "reason": "Qwen3.5 input-embedding unpacking in TRL and GPT-OSS tensor-parallel failures are unrelated." - }, - { - "left": "issue:43716", - "right": "issue:45685", - "accept": false, - "reason": "Mistral-3 dtype mismatch and MPS histogram limitation are different issues." - }, - { - "left": "issue:43828", - "right": "issue:44292", - "accept": false, - "reason": "Autocast dtype mismatch and Qwen NVFP4 runtime error are unrelated." - }, - { - "left": "issue:45507", - "right": "issue:45685", - "accept": false, - "reason": "GraniteMoEHybrid method error and MPS histogram not implemented are unrelated." - }, - { - "left": "issue:43716", - "right": "issue:44292", - "accept": false, - "reason": "Different model families and different failure modes." - }, - { - "left": "issue:44292", - "right": "issue:44918", - "accept": false, - "reason": "Qwen NVFP4 runtime error and TRL SFT embedding issue are unrelated." - }, - { - "left": "issue:43039", - "right": "issue:44918", - "accept": false, - "reason": "Liger kernel cross-entropy call issue and TRL SFT unpacking failure are different bugs." - }, - { - "left": "issue:41762", - "right": "issue:43638", - "accept": false, - "reason": "Both are ZeRO-3 loading/index errors, but they involve different model families and are not clearly the same concrete failure." - }, - { - "left": "issue:43827", - "right": "issue:44512", - "accept": false, - "reason": "Both are stale docs after v5 removals, but they refer to different removed APIs and pages." - }, - { - "left": "issue:43716", - "right": "issue:44560", - "accept": false, - "reason": "Mistral-3 dtype mismatch and Qwen3-VL video StopIteration are unrelated." - }, - { - "left": "issue:43828", - "right": "issue:45685", - "accept": false, - "reason": "Autocast dtype mismatch and MPS histogram limitation are unrelated." - }, - { - "left": "issue:44560", - "right": "issue:45685", - "accept": false, - "reason": "Qwen3-VL video processing and MPS histogram limitation are unrelated." + "reason": "Both mention loading/runtime failures, but the model and failure mechanisms differ." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Liger kernel cross-entropy behavior and autocast dtype mismatch are different issues." + "reason": "Different models and different numerical mismatch symptoms." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed collectives mismatch and compute_loss signature regression are unrelated." + "reason": "DeepSpeed collective mismatch vs trainer API signature mismatch are unrelated." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "FSDP activation-checkpointing error and GenerationMixin import regression are unrelated." + "reason": "Activation-checkpointing/FSDP issue vs import failure are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Liger kernel cross-entropy behavior and mask-shape IndexError are unrelated." + "reason": "Different subsystems and different concrete errors." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "TextClassificationPipeline docs mismatch and run-command docs removal are different doc issues." + "reason": "Both are docs issues, but they concern different deprecated API text." }, { "left": "issue:41762", "right": "issue:43596", "accept": false, - "reason": "Both are ZeRO-3 index-out-of-bounds reports, but they point to different models and likely different code paths." + "reason": "Both are ZeRO-3 loading errors, but the model contexts differ and the shared root cause is not established." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Gemma3 vLLM config error and Qwen3 auto-device-map CUDA assert are unrelated." + "reason": "Different models and different runtime failures." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "GPT-OSS AMD GPU issue and table-question-answering crash are unrelated." + "reason": "Unrelated model/backend failure vs table QA crash." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Voxtral processor failure and Gemma4 device mismatch are different issues." + "reason": "Processor failure vs device mismatch are different bugs." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger kernel cross-entropy behavior and Qwen3-VL video StopIteration are unrelated." + "reason": "Cross-entropy/Liger-Kernel issue vs video StopIteration are unrelated." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Same AutoTokenizer regression producing wrong token IDs across Granite and other model families." + "reason": "Same v5 AutoTokenizer token-ID regression; issue:45920 is the broader multi-model report while issue:45812 is Granite-specific." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Corrupted checkpoint and NemotronH loading failure are different problems." + "reason": "Checkpoint corruption vs load failure are different problems." }, { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "GLM-4.7-Flash load failure and Ovis2 checkpoint corruption are unrelated." + "reason": "Different models and unrelated failure modes." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "The mask-shape IndexError and the ZeRO-3 BERT zero-size index error are different failures." + "reason": "Same generic mask/index error wording, but not enough evidence of one underlying bug." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper model loading and table-question-answering crash are unrelated." + "reason": "Whisper model load/runtime issue vs table QA task crash are unrelated." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Same broad ZeRO-3 symptom family, but the model-specific repros are too different to merge confidently." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs cleanup, but they refer to different removed v5 surfaces." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Different model/runtime issues; no shared concrete bug." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "Different model families and unrelated behavior." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different model paths and unrelated backend errors." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Different bugs with different model contexts." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger-Kernel cross-entropy issue vs SFT embedding unpacking are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same ZeRO-3 Bert loading failure with the identical `index 0 is out of bounds for dimension 0 with size 0` error." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the accepted duplicate pairs are reasonably supported by the titles: 45812/45920, 45859/45902, and 43596/43638. The non-duplicate calls are generally conservative and do not overclaim beyond the evidence available in the issue titles." + "feedback": "Grounded and conservative overall. The accepted duplicate pairs are well-supported by the titles, and the canonical issue choice for the v5 AutoTokenizer regression is reasonable from the packet." }, "evaluator_used": true, "retried": false, @@ -546,8 +450,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -638,20 +542,20 @@ "issue:43645|issue:45399", "issue:44530|issue:45005" ], - "prepared_review_unit_hash": "3c0361675d0f6916b1ef2bf14c2b39f938db6f166e29a2bfbbda854a4950613e", + "prepared_review_unit_hash": "58bcb3482a849cba68ac270f6c09d4875e4c8c24d3438ab773479384c83fa1af", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 37, - "serialized_chars": 37923, + "serialized_chars": 37921, "estimated_input_tokens": 9481, "estimated_eval_tokens": 19218 }, @@ -660,244 +564,235 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers issues; only one soft edge looks like a plausible duplicate around custom-module loading/cache resolution. The rest are different models, different failure modes, or different code paths.", - "confidence": 0.44, - "canonical_issue_reason": "Issue 45020 is the broadest representative of the cluster\u2019s recurring theme of recent-version regressions breaking model loading/usage, especially around custom/remote code. It is still not a true duplicate of most items, but it is the best single issue to summarize the cluster.", + "summary": "Mostly unrelated Transformers issues are mixed together here; the soft links are generally just broad overlaps in model family, error type, or subsystem rather than the same concrete bug. I would not merge any of the soft pairs, and I\u2019d use the broad remote_code regression as the best representative issue.", + "confidence": 0.35, + "canonical_issue_reason": "issue:45020 is the broadest, most representative open bug report in the set: it describes a general regression where recent Transformers versions break models using `remote_code`, which is a common umbrella for several loading/configuration failures.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45020 is the strongest overall representative because it is broad, recent, and captures the widest shared regression pattern in the set. Most other issues are narrower model-specific breakages.", + "best_issue_reason": "issue:45020 is the strongest global representative because it is general, clearly scoped, and likely to cover a wider class of user-facing breakages than the more model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding-window behavior, but they affect different model families and describe different concrete failures." + "reason": "Both mention sliding-window related config issues, but they affect different model families and symptoms; not the same concrete bug." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "One is a Torch version compatibility problem; the other is a v5 pydantic/config field regression." + "reason": "One is a Torch version incompatibility, the other is a v5 Pydantic/config regression; too different to be duplicates." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "These are different incompatibility classes: a Zero-3/Bert index error versus a specific torch/transformers version mismatch." + "reason": "Different failure modes: DeepSpeed ZeRO-3 index error vs a Torch/Transformers compatibility issue." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Different symptom and subsystem: detokenization bug versus table-question-answering runtime crash." + "reason": "Completely different bugs: detokenization regression vs table-question-answering crash." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both involve Qwen3.5 and chat APIs, but the failures are on different model sizes and different execution paths." + "reason": "Both involve Qwen3.5 and chat paths, but the model size, API path, and failure mode differ; not clearly the same bug." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper loading failure and a PretrainedConfig/pydantic regression are unrelated." + "reason": "Whisper runtime failure is unrelated to the Pydantic field regression in Transformers v5.4.0." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "One is a model-load TypeError with init_empty_weights; the other is a task-specific crash in table QA." + "reason": "Different domains and code paths: Qwen NVFP4 loading vs table-question-answering crash." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "Different models and different bugs: weight tying in Mistral3 versus a corrupted Ovis2 checkpoint." + "reason": "Different models and symptoms; tied lm_head weights vs a corrupted checkpoint report." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Both are model-specific runtime issues, but they are distinct bugs in different architectures and code paths." + "reason": "Both are Gemma-related, but one is a missing config attribute and the other is a device mismatch at forward time." }, { "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Chat streaming failure and an MPS histogram/int issue in MoE are unrelated." + "reason": "Streaming chat failure on Qwen3.5 is unrelated to the MPS histogram/int error in a MoE model." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Grouped-mm tracing failure and an MPS integer-op failure are different concrete problems." + "reason": "Different subsystems and symptoms: torch dynamo/grouped_mm tracing error vs MPS unsupported histogram op." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both are dtype-related in MoE models, but they occur in different settings and are not the same failure mode." + "reason": "Both are dtype/runtime problems in MoE models, but the model families and exact code paths differ; too broad to merge." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Processor failure for Voxtral and a template-node compilation TypeError are unrelated." + "reason": "Processor failure for Voxtral is unrelated to a template-node compilation error." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "ImportError for GenerationMixin and a Gemma3 sliding_window attribute error are unrelated." + "reason": "Import path breakage for GenerationMixin is unrelated to the Gemma sliding_window_pattern issue." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are shape/index errors, but one is a mask mismatch and the other is a Zero-3/Gemma3 loading failure." + "reason": "Both are index/shape errors, but one is a mask-size mismatch and the other is a zero-size index during Gemma3 loading." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "A torch/transformers version incompatibility is not the same as a Gemma4 device-map mismatch." + "reason": "Version incompatibility and device placement mismatch are different bugs." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "Both concern model loading, but one is PaddleOCR-VL in vLLM and the other is GGUF support for gpt-oss." + "reason": "Different model-support gaps: PaddleOCR-VL loading in vLLM vs GGUF gpt-oss architecture support." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Missing pad_token_idx in one config class is not the same bug as a pydantic PretrainedConfig field regression." + "reason": "Missing `pad_token_idx` in StableLmConfig is a model config regression, not the same as the PretrainedConfig/Pydantic issue." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Int4 memory/OOM regression and tied-weights translation-model issues are different problems." + "reason": "Both are v5 regressions, but one is int4 memory/OOM behavior and the other is tied-weights handling on translation models." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Mixtral tracing failure and GPT-OSS tensor-parallel failure are different model/runtime issues." + "reason": "Different MoE-related problems: torch dynamo grouped_mm failure vs tensor-parallel support issue for GPT-OSS." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Autocast dtype mismatch and chat/completions streaming failure are unrelated." + "reason": "Different model families and error paths; dtype mismatch under autocast is not the same as a streaming chat API crash." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig pydantic breakage are different loader/config issues." + "reason": "init_empty_weights constructor argument failure is unrelated to the PretrainedConfig/Pydantic regression." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF architecture support and Qwen3.5 streaming inference are unrelated." + "reason": "GGUF architecture support and Qwen3.5 streaming inference failure are distinct issues." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Both mention recent versions, but one is remote_code breakage and the other is a torch/transformers dependency incompatibility." + "reason": "A general remote_code regression is not the same as a Torch 2.4.1/Transformers 5.8.0 incompatibility." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Sliding-window model behavior and a config-field regression are unrelated." + "reason": "Sliding-window config handling and Pydantic PretrainedConfig breakage are unrelated." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Mask-shape mismatch and Zero-3/Bert index-out-of-bounds are different runtime failures." + "reason": "Both are indexing errors, but the triggering conditions and code paths differ substantially." }, { "left": "issue:43541", "right": "issue:44589", "accept": false, - "reason": "Grouped_mm tracing and missing Float8 storage are unrelated errors." + "reason": "Grouped_mm tracing failure and missing Float8 storage object are different low-level runtime issues." }, { "left": "issue:43054", "right": "issue:43994", "accept": false, - "reason": "Same model family, but the reports describe different symptoms and likely different root causes." + "reason": "Both concern Siglip2 quality, but one is a text-embedding regression and the other is a generic nonsensical-output report; not enough to treat as the same bug." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "TensorBoard-without-TensorFlow backend error is unrelated to torch/transformers incompatibility." + "reason": "TensorBoard-without-TensorFlow backend detection is unrelated to Torch/Transformers version incompatibility." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite-config unwrapping and a meta-to-empty memory initialization bug are different code paths." + "reason": "Different bugs: config unwrapping for composite multimodal models vs uninitialized rotary embedding memory after meta/to_empty materialization." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "High perplexity on gpt-oss and a GLM5 issue are not the same underlying bug." + "reason": "High perplexity on gpt-oss and a GLM5 issue are not the same underlying problem." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP 2-D weight error and TRL embedding unpacking failure are unrelated." + "reason": "FSDP weight-shape error and TRL SFT embedding unpacking failure are different training-path bugs." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Image preprocessor dtype mismatch and Qwen3.5 SFT unpacking failure are different bugs." + "reason": "Mistral image dtype mismatch is unrelated to Qwen3.5 embedding unpacking in TRL." }, { "left": "issue:45632", "right": "issue:45698", - "accept": true, - "reason": "Both point to the same custom-module loading / trust_remote_code cache-resolution problem, where the wrong module can be resolved from a colliding path." + "accept": false, + "reason": "Both involve remote/custom loading, but one is a cache-path collision and the other is loading the wrong custom module after save/load; not clearly the same bug." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Both involve tied weights, but they are on different model families and likely different fixes." - }, - { - "left": "issue:43645", - "right": "issue:45399", - "accept": false, - "reason": "Jupyter custom-model initialization and flash-attn2 fallback gating are unrelated." + "reason": "Both mention tied weights, but they concern different models and different manifestations of the problem." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and tied-weights translation-model issues are unrelated." + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight handling on translation models." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: most titles do look like distinct model-/runtime-specific regressions, and the lone accepted soft edge (45632/45698) is a plausible custom-module loading duplicate. The other soft-edge rejections are conservative and do not overstate duplication evidence. " - }, - "evaluator_used": true, + "evaluator_result": null, + "evaluator_used": false, "retried": false, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -909,8 +804,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -999,247 +894,247 @@ "issue:43881|issue:43883", "issue:41762|issue:42915" ], - "prepared_review_unit_hash": "1853eb6d2d1b103e7b5fbc4f0bbb8f7638043c8625357e8fc90231dcecd18c93", + "prepared_review_unit_hash": "988bfdee93c0550ba54fc2e49ba73a7cbb96493a04ea0b1f75ab3a9218ba7c10", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37476, - "estimated_input_tokens": 9369, - "estimated_eval_tokens": 18994 + "serialized_chars": 37480, + "estimated_input_tokens": 9370, + "estimated_eval_tokens": 18996 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly unrelated issues, but the non-persistent-buffer regression is a small duplicate family rather than a single pair: issue:43644, issue:43950, and issue:44534 all describe the same Transformers v5 buffer-loading corruption bug. I would anchor the cluster on issue:43950 as the clearest representative.", + "summary": "Mostly a grab-bag of unrelated Transformers issues; only one soft pair looks like a true duplicate: the non-persistent buffer corruption regression in v5. The rest share broad themes like loading, quantization, or model-specific failures but not the same concrete bug.", "confidence": 0.79, - "canonical_issue_reason": "issue:43950 is the clearest and most specific report of the shared Transformers v5 regression, and it sits at the center of the duplicate family with issue:43644 and issue:44534.", + "canonical_issue_reason": "issue:43950 is the best canonical issue because it states the core regression clearly and broadly: `from_pretrained()` corrupting non-persistent buffers in Transformers 5.x. It is the most representative of the one clear duplicate pair with issue:44534.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43950 is the best representative issue because it precisely names the regression (`from_pretrained()` corrupting non-persistent buffers) and matches the duplicate family in the cluster.", + "best_issue_reason": "issue:43950 is the strongest standalone issue in this set: concrete, well-scoped, and it matches a near-identical duplicate entry.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "Different topics: a discussion about DSA indexing vs docs mentioning a removed CLI command." + "reason": "Different problems: a discussion about DSA indexer behavior vs docs referencing a removed run command." }, { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "Both involve model/runtime compatibility, but they describe different failure modes and code paths." + "reason": "Both involve model/runtime behavior, but one is a Qwen3Moe FP8 failure and the other is a flash-attn fallback gating issue." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization/offloading behavior and tied-weights handling are unrelated bugs." + "reason": "Quantization/offloading behavior is unrelated to tied-weights handling in translation models." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint loading and translation-model tied-weights regressions are different problems." + "reason": "Model checkpoint loading failure is not the same underlying bug as tied-weights handling in v5." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A typo in cache strategies is not the same issue as obsolete docs for the v5 run command." + "reason": "A cache-strategy typo is unrelated to the docs issue about a removed command." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different models and failures: Qwen3.5 embedding unpacking vs GraniteMoEHybrid invalid method call." + "reason": "Different code paths: Qwen3.5 embedding unpacking in TRL vs GraniteMoEHybrid calling an invalid method." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Detokenization bug and NVFP4 runtime error do not look like the same underlying defect." + "reason": "Both are model-specific loading/tokenization failures, but they are distinct models and distinct failure modes." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "These describe the same Transformers v5 regression: non-persistent buffers are filled/corrupted with junk during load." + "reason": "Near-identical regression: both report Transformers v5 filling non-persistent buffers with junk / corrupting `register_buffer(persistent=False)` state." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "Different model-specific runtime errors with no shared code-path evident." + "reason": "Different failures: Qwen NVFP4 runtime error vs GraniteMoEHybrid invalid method call." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "SigLIP2 docs-example errors and SigLIP2 embedding-quality concerns are related by model family only, not the same bug." + "reason": "SigLIP2 example/documentation errors are not the same as degraded text embeddings." }, { "left": "issue:43643", "right": "issue:43950", "accept": false, - "reason": "`trust_remote_code` missing fields and non-persistent buffer corruption are distinct issues." + "reason": "Remote-code config field loss is a different bug than non-persistent buffer corruption." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Both mention Qwen3.5/training, but one is a dtype mismatch under autocast and the other is an embedding-unpacking trainer failure." + "reason": "Both mention dtype/position_ids-related problems, but they occur in different models and different failure mechanisms." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Trainer `compute_loss` signature mismatch is unrelated to Gemma3 ZeRO-3 loading failure." + "reason": "Trainer API keyword mismatch is unrelated to Gemma3 ZeRO-3 loading failure." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "Corrupted checkpoint loading and noisy load logs are not the same underlying bug." + "reason": "Checkpoint corruption and loading noise are not the same underlying issue." }, { "left": "issue:41628", "right": "issue:45800", "accept": false, - "reason": "AutoImageProcessor import failure and torch/transformers version incompatibility are different compatibility problems." + "reason": "Import/export availability problems are unrelated to torch/Transformers version incompatibility." }, { "left": "issue:43950", "right": "issue:45902", "accept": false, - "reason": "Both involve buffer/state materialization, but one is a v5 load corruption regression while the other is a Qwen3_5Moe rotary embedding initialization bug." + "reason": "Both concern buffer/initialization state, but one is a general regression and the other is a specific `inv_freq` uninitialized-memory bug." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "bitsandbytes parameter-constructor incompatibility is unrelated to missing Float8 storage support." + "reason": "Bitsandbytes constructor incompatibility is unrelated to missing float8 storage support." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "Both concern SigLIP2, but one is a quality regression and the other is a tokenizer-version behavior change." + "reason": "Worse SigLIP2 text embeddings are not the same as version-dependent tokenization changes." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to Molmo tied-weights attribute failure." + "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to missing `all_tied_weights_keys` on Molmo." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Unexpected zero probabilities and worse text embeddings are similar only at a high level, not the same concrete bug." + "reason": "Unexpected zero probabilities in SigLIP2 and worse SigLIP2 text embeddings are related to the model family but are different bugs." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "Both are Qwen3.5 attention/rotary issues, but one is flash-attn illegal memory access and the other is CPU offload incompatibility." + "reason": "Both touch Qwen3.5/MoE, but one is Flash Attention misinterpreting 3D position_ids and the other is CPU offload compatibility." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Gemma4 FSDP2 attention key error and Qwen3_5Moe rotary buffer initialization are unrelated." + "reason": "Gemma4 FSDP2 shared_kv_state KeyError is unrelated to Qwen3.5 rotary embedding uninitialized memory." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Different failures: tokenizer loading error vs model loading failure for a different checkpoint." + "reason": "Bad tokenizer loading error and a model loading failure are not the same defect." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "LoRA gradient checkpointing ineffectiveness and int4 CUDA reserved-memory OOM are separate training/memory issues." + "reason": "Gradient checkpointing with LoRA is unrelated to int4 quantization memory regression." }, { "left": "issue:43054", "right": "issue:43493", "accept": false, - "reason": "Both concern SigLIP2, but one is a quality regression and the other is a HF-vs-JAX implementation discrepancy." + "reason": "Both are SigLIP2-related, but one is text embedding quality and the other is HF-vs-JAX implementation discrepancy." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Generic loading noise and inability to load a specific model are not the same bug." + "reason": "Generic load-time noise vs a specific failure to load one model are not the same bug." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "Missing fields from `AutoConfig` with remote code and broader remote-code breakage are related only thematically." + "reason": "Remote-code field loss is not the same as broad recent version breakage for remote_code models." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Detokenization regression and Python 3.13 flash_attn import error are unrelated." + "reason": "Detokenization bug and import-utils flash_attn KeyError are unrelated." }, { "left": "issue:43975", "right": "issue:45237", "accept": false, - "reason": "Detokenization bug and AMD GPU support issue are different subsystems and failure modes." + "reason": "Detokenization on DeepSeek Coder is unrelated to GPT-OSS-20B AMD GPU support." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Loading noise and a specific model load failure are not duplicates." + "reason": "Both are loading-related, but one is a generic noise complaint and the other is a concrete model load failure." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "Both are training/memory-related, but one is checkpointing effectiveness and the other is MoE memory usage." + "reason": "LoRA gradient checkpointing and Qwen3 MoE memory usage are different training-path issues." }, { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Qwen3VL weight-only loading and GLM5 are different model-specific issues." + "reason": "Qwen3VL weight_only load error is unrelated to GLM5." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Two separate model-loading errors for different models and different root causes." + "reason": "Different models and different failure causes: GLM-4V loading vs Molmo tied-weights attribute missing." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Gemma3 ZeRO-3 loading and Qwen3Moe FP8 config failures are unrelated." + "reason": "Gemma3 ZeRO-3 loading failure and Qwen3Moe FP8 failure are distinct model-specific issues." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The summary is mostly grounded, but it overstates the duplicate evidence around the non-persistent-buffer regression. In the packet, issue:43644 is also essentially the same bug as issue:43950 and issue:44534, so it is not accurate to call 43950/44534 the \"only strong duplicate pair.\" I\u2019d want the cluster description to reflect a small duplicate family rather than a single pair. Most soft-edge verdicts look conservative, but I can\u2019t accept the overstatement in the core summary." + "accept": true, + "feedback": "Grounded overall. The selected canonical issue is well supported by the packet, and the one accepted soft pair (43950/44534) is a plausible duplicate based on closely matching titles. The remaining soft-edge rejections are conservative and do not overclaim equivalence." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1341,226 +1236,226 @@ "issue:45325|issue:45902", "issue:43366|issue:43828" ], - "prepared_review_unit_hash": "084d0e989fd8f588e767d45c0eda3bfaae6c388d1ea2af041cd092dbf761de3d", + "prepared_review_unit_hash": "b42619d9cf91349ce6952ccd555bd4da59b9225c256f799424224dcc2af123a7", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 38069, - "estimated_input_tokens": 9518, - "estimated_eval_tokens": 19292 + "serialized_chars": 38066, + "estimated_input_tokens": 9517, + "estimated_eval_tokens": 19290 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous, but there are a couple of clear duplicate subclusters: one around v5 docs still referencing removed pipeline tasks, and one around the unexpected `_is_hf_initialized` argument during model loading. Most other pairs are only broadly similar by subsystem and should stay separate.", + "summary": "Mostly unrelated issues; a few soft duplicates stand out around docs, initialization compatibility, DeepSpeed ZeRO-3 loading, mask-shape IndexErrors, and Qwen video regressions.", "confidence": 0.67, - "canonical_issue_reason": "Issue 44509 is the clearest representative of the strongest duplicate subcluster here: it explicitly covers the v5 docs regression about removed pipeline tasks, and it is broader/more complete than 43827.", + "canonical_issue_reason": "issue:45020 is the broadest, most representative current-regression report in the set, and it provides a reasonable anchor for comparing the more model-specific loading/tokenizer failures.", "canonical_pr_reason": null, - "best_issue_reason": "44509 is the best single issue to anchor the docs-related subcluster because it states the regression most comprehensively and matches the paired docs report most directly.", + "best_issue_reason": "Open, high-signal umbrella issue covering recent transformer-version regressions with real user impact; broader and more reusable as a triage anchor than the narrow per-model tickets.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "Different bugs: MPS histogram failure on Int vs Gemma4 PLE device-mismatch during forward." + "reason": "Different symptoms and code paths: MPS histogram/int handling vs Gemma4 PLE device placement." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Both involve Qwen/LoRA warnings or compatibility, but the concrete failures differ and do not describe the same bug." + "reason": "Both mention Qwen/LoRA, but one is a version-compatibility load failure and the other is a config warning; not the same bug." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "ImportError for a missing class vs bitsandbytes Int8Params keyword-arg failure; not the same code-path." + "reason": "Importing a missing class and a bitsandbytes `_is_hf_initialized` constructor error are unrelated failures." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Broad remote_code regression report vs a specific Gemma4 device_map bug; too different to merge." + "reason": "Remote-code breakage is a different problem from a Gemma4 device mismatch in forward." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Tokenizer mapping addition failure is not the same bug as general remote_code breakage." + "reason": "Tokenizer-mapping guard failure is not the same as the broader remote_code regression." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading regression and flash-attn fallback logic are unrelated failure modes." + "reason": "Tokenizer loading failure for est-roberta is unrelated to flash-attn fallback gating." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "dtype mismatch under autocast is unrelated to Gemma4 device placement." + "reason": "Different model/runtime failures: autocast dtype mismatch vs device mismatch." }, { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "Same docs regression: both report v5 docs still referencing pipeline tasks that were removed." + "reason": "Same documentation regression: both still reference pipeline tasks removed in v5." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Mistral tokenizer regex patch crash vs flash-attn fallback gating; different issues." + "reason": "Tokenizer regex patch crash is unrelated to flash-attn fallback selection." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "General remote_code regression and Kimi-K2.5 tokenizer codec regression are distinct." + "reason": "Broad remote_code regressions and Kimi tokenizer codec handling are different root causes." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Weight tying warning in Mistral3 is unrelated to non-persistent buffer corruption in from_pretrained()." + "reason": "Weight tying in Mistral3 and non-persistent buffer corruption are distinct from each other." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "Loading failure for one model is not the same as flash-attn fallback gating." + "reason": "MobileLLM key error is unrelated to flash-attn fallback checks." }, { "left": "issue:42915", "right": "issue:45020", "accept": false, - "reason": "Gradient-checkpointing/LoRA behavior is unrelated to remote_code loading regressions." + "reason": "Qwen3_moe LoRA compatibility between branches is a different issue from remote_code breakage." }, { "left": "issue:43257", "right": "issue:43866", "accept": false, - "reason": "Qwen3 MoE weight conversion/loading with deepspeed is unrelated to a corrupted checkpoint." + "reason": "DeepSpeed weight conversion for Qwen3 MOE and a corrupted Ovis2 checkpoint are unrelated." }, { "left": "issue:43638", "right": "issue:45823", "accept": false, - "reason": "Zero3 index-out-of-bounds on a Bert model is not the same as Gemma4 PLE device mismatch." + "reason": "Zero3/Bert empty-weights IndexError is not the same as Gemma4 device placement." }, { "left": "issue:43792", "right": "issue:44220", - "accept": true, - "reason": "Both point to the same Whisper audio feature-extraction failure path around `_torch_extract_fbank_features()`." + "accept": false, + "reason": "Whisper not running is too broad to treat as the same bug as `_torch_extract_fbank_features()` without a clearer shared traceback." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same underlying `_is_hf_initialized` TypeError during model loading/init_empty_weights; 43872 is the bitsandbytes manifestation." + "reason": "Same `_is_hf_initialized` incompatibility triggered during model init/loading." }, { "left": "issue:43940", "right": "issue:45313", - "accept": false, - "reason": "Both are DeepSpeed ZeRO-3 loading failures, but they affect different model families and report different missing-weight behavior." + "accept": true, + "reason": "Both report DeepSpeed ZeRO-3 failing to load Qwen3-family weights with missing parameters." }, { "left": "issue:42915", "right": "issue:43872", "accept": false, - "reason": "Gradient checkpointing with LoRA is unrelated to the `_is_hf_initialized` loading error." + "reason": "LoRA/Qwen3_moe branch compatibility is unrelated to the bitsandbytes/init_empty_weights constructor error." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Bad AutoTokenizer error message for Voxtral vs tokenizer loading regression for EMBEDDIA/est-roberta are different reports." + "reason": "A bad error message for Voxtral tokenization is not the same as an actual tokenizer load failure for a different model." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "Unit-test model loading failure for GLM-4.7-Flash and NemotronH checkpoint-loading failure are separate model-specific bugs." + "reason": "Different model families and different loading failures." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Unexpected `num_items_in_batch` kwarg in compute_loss is unrelated to a mask/tensor shape mismatch." + "reason": "Unexpected trainer kwargs and mask/tensor shape mismatch are unrelated." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Both mention DeepSpeed/Qwen, but one is missing weight conversion and the other is gradient averaging under GAS; different defects." + "reason": "Both involve DeepSpeed, but one is weight conversion and the other is gradient averaging under GAS." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing ineffective under PEFT LoRA is not the same as a missing supports_gradient_checkpointing flag." + "reason": "Gradient checkpointing ineffective in PEFT LoRA is not the same as a missing supports_gradient_checkpointing flag on a trust_remote_code model." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "RLHF gradient explosion and flash_attention s_aux=None crash are unrelated." + "reason": "NaN/SDPA fallback behavior in RLHF is unrelated to the flash_attention `s_aux=None` crash." }, { "left": "issue:41093", "right": "issue:44805", - "accept": false, - "reason": "These are both mask-shape IndexErrors, but the reported shapes and contexts are distinct; not enough to treat as the same bug." + "accept": true, + "reason": "Same mask-vs-index length mismatch IndexError, just reported with different concrete sizes." }, { "left": "issue:43872", "right": "issue:45538", "accept": false, - "reason": "Loading/init_empty_weights TypeError is unrelated to CLIPTokenizer max-length behavior." + "reason": "bitsandbytes init compatibility and CLIPTokenizer max length are unrelated." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Different tokenizer-loading reports for different models and error surfaces." + "reason": "Voxtral error-message quality and sleng-bert loading failure are different tokenizer issues." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "PagedAttentionCache linear_attention crash is unrelated to NemotronH checkpoint loading." + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." }, { "left": "issue:44479", "right": "issue:44560", - "accept": false, - "reason": "Video-input regression across several Qwen variants is not the same as a StopIteration in Qwen3-vl-embedding." + "accept": true, + "reason": "Both describe the same Qwen video-input regression in 5.3.0; 44560 is a specific manifestation." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed training weight-dimension error is unrelated to void segmentation map processing." + "reason": "DeepSpeed training weight-shape error and void segmentation-map processing are unrelated." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "RoPE position_id scaling bug and uninitialized-memory after to_empty() are different implementation bugs." + "reason": "Qwen2.5-VL rope scaling and Qwen3.5-MoE uninitialized-memory materialization are different bugs." }, { "left": "issue:43366", @@ -1572,7 +1467,7 @@ }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The two accepted duplicate subclusters are well supported by the issue titles: the v5 docs/pipeline-task regression and the `_is_hf_initialized` loading/init_empty_weights TypeError. The remaining pair verdicts are appropriately cautious and do not overclaim sameness from loose subsystem overlap." + "feedback": "The summary is grounded in the packet and stays conservative. The accepted soft-edge pairs are plausible duplicates or near-duplicates (docs v5 pipeline references, `_is_hf_initialized` init compatibility, ZeRO-3 weight loading failures, mask-shape IndexErrors, and the Qwen video regression), while the rejected pairs mostly distinguish different symptoms or root causes appropriately. No obvious overstatement of evidence." }, "evaluator_used": true, "retried": false, @@ -1588,8 +1483,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -1666,8 +1561,8 @@ "issue:43317|issue:44756", "issue:43644|issue:44534", "issue:38175|issue:39692", - "issue:42491|issue:43720", "issue:44960|issue:44977", + "issue:42491|issue:43720", "issue:45313|issue:45357", "issue:44451|issue:44488", "issue:41720|issue:42915", @@ -1676,20 +1571,20 @@ "issue:42915|issue:44756", "issue:45488|issue:45812" ], - "prepared_review_unit_hash": "52ad9aaf453cf97dec573a2163e1bab11295d03396238614daa879c8a652c36c", + "prepared_review_unit_hash": "6d5b54cadeeb026b9d33571c5677d245041865d7644ec0ebff17e4839df2c973", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37261, + "serialized_chars": 37263, "estimated_input_tokens": 9316, "estimated_eval_tokens": 18888 }, @@ -1698,240 +1593,210 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers issues, with a few true duplicate pairs around v5 regressions, SigLIP2 output corruption, and BERT/model-loading failures.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43644 is the clearest representative of the v5 non-persistent-buffer corruption regression and has a strong duplicate in 44534; it is the cleanest anchor among the accepted issue pairs.", + "summary": "This cluster is mostly heterogeneous Transformers regressions. I only found one clear duplicate pair (the v5 non-persistent-buffer junk regression) and one near-identical custom-model-init pair; the rest are too loosely related to merge.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 43644 is the clearest, most specific representative of the shared non-persistent-buffer regression and is better framed than its duplicate counterpart.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43644 is the best single issue to represent the cluster\u2019s most concrete, high-signal regression theme without over-generalizing across unrelated model-specific bugs.", + "best_issue_reason": "As a local anchor for the one unambiguous duplicate subcluster, issue 43644 is the strongest fit; the broader cluster is too mixed for a single global canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4 problems, but one is an FSDP2 KeyError and the other is a device-mismatch forward failure; different bugs/code paths." + "reason": "Same model family, but the bugs are different: FSDP2 KeyError in attention vs. device-mapping forward mismatch." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Different models and failure modes: AudioFlamingo3 track leakage vs Whisper empty-transcription behavior." + "reason": "Different code paths: batched AudioFlamingo token/embedding leakage vs Whisper empty-transcription behavior after align_special_tokens." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "Unrelated: GGUF gpt-oss support request vs Qwen3-MoE sliding_window issue." + "reason": "Both mention Qwen/GGUF, but one is about gpt-oss architecture support and the other is a sliding_window bug." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Different bugs: non-persistent buffer corruption vs Kosmos2.5 long-OCR index error." + "reason": "Unrelated symptoms: non-persistent buffer corruption vs Kosmos2.5 OCR index error." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Different model/code paths: Bamba torch.compile+SDPA failure vs Qwen3.5 flash-attention crash from 3D position_ids handling." + "reason": "Different failures: Bamba torch.compile SDPA vs Qwen3.5 flash-attention position_ids misinterpretation." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Import error for a missing Qwen2.5-VL symbol is unrelated to meta-device loading regressions." + "reason": "Import error for Qwen2.5-VL is unrelated to the meta-device loading regression." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same underlying regression: Transformers v5 breaking custom model initialization, with the notebook report just a narrower variant." + "reason": "These look like the same Transformers 5 custom-model initialization regression; one is the notebook-specific framing of the same failure." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "No shared bug; gpt-oss support is unrelated to the Float8 storage-object error." + "reason": "Different problems: GGUF gpt-oss support vs missing Float8 storage type during loading." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3Moe FP8 failure and meta-device model loading are distinct issues." + "reason": "Qwen3Moe FP8 failure and meta-device loading are separate bugs." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Different symptoms and contexts: Qwen3-vl video StopIteration vs TRL embedding-unpacking failure." + "reason": "Both are Qwen3.5/VL-related, but one is a video StopIteration issue and the other is TRL SFT embedding unpacking." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "3d_parallel.py runtime failure is unrelated to GGUF gpt-oss support." + "reason": "3d_parallel.py runtime failure is not the same bug as GGUF gpt-oss support." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import error is unrelated to Qwen3Moe FineGrainedFP8Config failure." + "reason": "Importing GenerationMixin and Qwen3Moe FP8 are unrelated regressions." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer-mapping lookup bug and Gemma4 device mismatch are unrelated." + "reason": "Tokenizer-mapping add-new-model-like failure is unrelated to Gemma4 device mismatch." }, { "left": "issue:38175", "right": "issue:43994", - "accept": true, - "reason": "Same model family and same symptom class: SigLIP2 producing incorrect/degenerate outputs in inference." + "accept": false, + "reason": "Same model name, but the titles only show broad bad-output symptoms; not enough to confirm the same concrete bug path." }, { "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized flag breakage is unrelated to Kosmos2.5 OCR indexing." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different failure modes: video StopIteration vs mask/tensor shape IndexError." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Strict config loading and glm4v config validation are separate issues." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "Zero-3/BertModel index error is not the same as the tokenizer-mapping add-new-model-like bug." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Both involve loading models, but the specific failures differ too much to treat as duplicates." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer mapping failure and mask-shape IndexError are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "gpu+cpu offload dequantized-model loading is a different issue from disabling mmap on Strix Halo." + }, + { + "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Same Transformers v5 regression about non-persistent buffers being filled with junk." + "reason": "Near-verbatim duplicate: both report the same Transformers 5 non-persistent-buffer junk regression." }, { - "left": "issue:42491", - "right": "issue:43720", + "left": "issue:38175", + "right": "issue:39692", "accept": false, - "reason": "LoRA compatibility with qwen3_moe on hf4.x is unrelated to BitNet packed-weight unpacking during accelerate loading." + "reason": "Both mention SigLIP2, but one is output corruption and the other is documentation example errors/quantization failure." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "GLM5 model thread and Qwen3.5 flash-attention crash are not the same bug." + "reason": "GLM5 and Qwen3.5 flash-attention are different model-specific bugs." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "LoRA qwen3_moe hf4.x vs BitNet packed-weight unpacking are unrelated regressions." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "ZeRO-3 weight loading for language_model and save_pretrained visual-encoder key corruption are different code paths." + "reason": "DeepSpeed ZeRO-3 language_model weight loading and incorrect visual encoder keys are different Qwen3.5 problems." }, { "left": "issue:44451", "right": "issue:44488", - "accept": true, - "reason": "Both reports describe the same inability to load specific BERT models in the current version." + "accept": false, + "reason": "Both are loading failures, but the titles do not establish the same concrete root cause." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Qwen3 auto-device-map cuda assert and Qwen3Moe FP8 failure are unrelated." + "reason": "A800 cudaErrorAssert on auto device mapping is unrelated to Qwen3Moe FP8 failure." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "qwen3_moe LoRA/HF version issue is unrelated to Gemma4 resize_token_embeddings behavior." + "reason": "LoRA checkpoint compatibility and gemma4 resize_token_embeddings behavior are different bugs." }, { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Trainer resume/max_steps logic and SGD argument handling are unrelated." + "reason": "Trainer resume/max_steps accounting and SGD args not being used are separate training issues." }, { - "left": "issue:45488", - "right": "issue:45812", + "left": "issue:42915", + "right": "issue:44756", "accept": false, - "reason": "Tokenizer override for DeepSeek family and Granite token-ID regression are different bugs." - }, - { - "left": "issue:43366", - "right": "issue:44589", - "accept": false, - "reason": "GGUF support for gpt-oss is unrelated to the Float8 storage lookup error." - }, - { - "left": "issue:38175", - "right": "issue:39692", - "accept": false, - "reason": "Both mention SigLIP2, but one is an inference-output bug and the other is a documentation example with model/processor mismatch and quantization issues." - }, - { - "left": "issue:43596", - "right": "issue:44661", - "accept": false, - "reason": "DeepSpeed ZeRO-3/Bert initialization failure is unrelated to add-new-model-like tokenizer mapping behavior." - }, - { - "left": "issue:43638", - "right": "issue:44661", - "accept": false, - "reason": "Non-pretrained Bert ZeRO-3 index error is unrelated to the tokenizer-mapping bug." - }, - { - "left": "issue:44560", - "right": "issue:44805", - "accept": false, - "reason": "Qwen3-vl video StopIteration and a generic mask-shape IndexError are different failure modes." - }, - { - "left": "issue:44877", - "right": "issue:45030", - "accept": false, - "reason": "Strict config rejecting granite_speech and glm4v validation changes are unrelated." - }, - { - "left": "issue:43317", - "right": "issue:44756", - "accept": false, - "reason": "Dequantized-model offload failure on device_map=auto is unrelated to the Strix Halo mmap OOM report." - }, - { - "left": "issue:43644", - "right": "issue:44534", - "accept": true, - "reason": "Exact same v5 regression: non-persistent buffers are being corrupted/junk-filled on load." - }, - { - "left": "issue:43645", - "right": "issue:43646", - "accept": true, - "reason": "Same root bug: custom model initialization breaks in Transformers v5, with one report focused on notebooks." - }, - { - "left": "issue:42491", - "right": "issue:43720", - "accept": false, - "reason": "Different models and mechanisms; not the same underlying bug." - }, - { - "left": "issue:44960", - "right": "issue:44977", - "accept": false, - "reason": "No concrete overlap beyond being recent model-specific reports." - }, - { - "left": "issue:45313", - "right": "issue:45357", - "accept": false, - "reason": "Save/load regressions in different subsystems are not mergeable into one bug fix." - }, - { - "left": "issue:44451", - "right": "issue:44488", - "accept": true, - "reason": "Both are duplicate load failures for similar BERT models under the current version." + "reason": "Qwen3Moe FP8 and Strix Halo mmap OOM are unrelated." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Tokenizer-pretokenizer override regression and Granite token-ID corruption are not the same defect." + "reason": "Tokenizer pre-tokenizer override regression and Granite AutoTokenizer wrong IDs are different tokenizer issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays reasonably conservative: most pair rejections are supported by distinct model/symptom titles, and the accepted pairs are plausible duplicates or near-duplicates (especially the v5 non-persistent-buffer regressions and the SigLIP2/BERT loading issues). The only slightly weaker accept is 44451/44488, but it is still defensible from the issue titles." + "feedback": "Grounded overall: the cluster is indeed mostly heterogeneous, and the one clear duplicate pair (43644/44534) is strongly supported by the titles. The custom-model-init pair (43645/43646) is also a reasonable conservative accept from the available evidence. No PR-pair mergeability issue applies here." }, "evaluator_used": true, "retried": false, @@ -1947,8 +1812,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -2007,24 +1872,24 @@ "issue:43257|issue:43854", "issue:44488|issue:44991", "issue:42491|issue:43931", - "issue:43856|issue:43873", "issue:43644|issue:43645", + "issue:43856|issue:43873", "issue:45362|issue:45406", "issue:43856|issue:44387", "issue:44910|issue:45588", "issue:42491|issue:45292", "issue:36010|issue:39404", "issue:45276|issue:45335", - "issue:43404|issue:43454", "issue:44164|issue:44291", + "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", @@ -2039,20 +1904,20 @@ "issue:44164|issue:44756", "issue:42947|issue:43381" ], - "prepared_review_unit_hash": "8edf06a84d9893475c180b2db81c05a3258cbcb91bcd4880f4300c5f9bd9998d", + "prepared_review_unit_hash": "3b3dc9fc5d6d43c85d50d4fcbc36da64c3e74b472910bfb9fca69a83f083567b", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 36, - "serialized_chars": 38613, + "serialized_chars": 38614, "estimated_input_tokens": 9654, "estimated_eval_tokens": 19564 }, @@ -2061,234 +1926,234 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated, model-specific bug reports rather than duplicates. The only convincing soft duplicate is the resize_token_embeddings/output_embeddings regression pair around Gemma4/general embedding propagation.", - "confidence": 0.82, - "canonical_issue_reason": "issue:45292 is the best canonical representative for the resize_token_embeddings/output_embeddings bug because it is the broader, model-agnostic report; issue:45276 reads like a Gemma4-specific instance of the same missing update path.", + "summary": "This cluster is mostly a loose topical grouping of unrelated Transformers issues. The only clear duplicate-like group is the resize_token_embeddings regression affecting Gemma4/T5Gemma embedding modules; the rest should stay separate.", + "confidence": 0.77, + "canonical_issue_reason": "issue:45276 is the best canonical issue for the duplicate set because it describes the broader resize_token_embeddings failure and already covers multiple affected embedding paths (embed_tokens_per_layer and output_embeddings).", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 is the strongest issue candidate overall: clear, generic title, same underlying bug as its close variant, and broad enough to subsume the duplicate report.", + "best_issue_reason": "issue:45276 is the strongest representative issue in this cluster: concrete repro, broader scope than the other resize_token_embeddings reports, and it best captures the shared underlying bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Both are model-loading failures, but for different models and failure modes (GLM-4.7-Flash test load vs Qwen3-VL shape mismatch). Not the same bug." + "reason": "Both are model-loading failures, but they affect different models and failure modes; no evidence of the same underlying bug or a single fixable code path." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2/vision models, but one is a docs/example error report and the other is an AutoModel/pipeline inference correctness issue." + "reason": "Both mention SigLIP2, but one is a documentation/example problem while the other is an inference correctness issue; not the same bug." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed is unrelated to GLM-4.7-Flash loading." + "reason": "Different models and different symptoms (Qwen3 MoE conversion/loading vs GLM-4.7 unit-test loading failure)." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Different models and different tokenizer-loading failures; too broad to be the same bug." + "reason": "Both are tokenizer-loading regressions, but for different models and likely different tokenizer/config breakages; too broad to merge." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "Both are Qwen-related loading problems, but one is an hf4.x/hf5.x LoRA compatibility issue and the other is a Qwen3-VL weight-shape mismatch." + "reason": "Distinct Qwen3 MoE training/load issues with different triggers and error classes." }, { - "left": "issue:43856", - "right": "issue:43873", + "left": "issue:43644", + "right": "issue:43645", "accept": false, - "reason": "Training memory inefficiency in Qwen3 MoE is not the same as offloading behavior with quantization." + "reason": "Both are Transformers 5.0.0 regressions, but one is about non-persistent buffers and the other about custom model init in Jupyter; not the same bug." }, { - "left": "issue:43644", - "right": "issue:43645", + "left": "issue:43856", + "right": "issue:43873", "accept": false, - "reason": "Both are Transformers 5.0.0 regressions, but one is junk non-persistent buffers and the other is Jupyter custom-model initialization." + "reason": "Both involve memory/quantization/offloading concerns, but the concrete failure modes and code paths differ." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different model families and different crashes: Qwen3.5 chat failure vs Gemma4Processor missing _tokenizer in serve." + "reason": "Different models and different failures (Qwen3.5 chat crash vs Gemma4Processor missing attribute)." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both are memory-related, but one is Qwen3 MoE training efficiency and the other is int4 quantization reserved-memory/OOM behavior." + "reason": "Both are memory-related, but one is training efficiency and the other is int4 quantization reserved-memory OOM; not the same bug." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Both involve attention code, but the bugs are different: Qwen3.5 Flash Attention position_ids handling vs sink-less-models flash_attention.py s_aux=None crash." + "reason": "Different attention/backend crashes with different causes and affected paths." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "LoRA hf4.x/hf5.x incompatibility is unrelated to resize_token_embeddings propagation." + "reason": "Both mention related Qwen/Gemma-style regressions, but they are not the same underlying bug or code path." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "ImportError for GenerationMixin and Whisper return_language pipeline regression are different issues." + "reason": "ImportError for GenerationMixin is unrelated to Whisper pipeline return_language behavior." }, { "left": "issue:45276", "right": "issue:45335", - "accept": false, - "reason": "Both concern resize_token_embeddings, but they hit different architectures and different internal embeddings; not clearly one fixable code-path." + "accept": true, + "reason": "Both report the same resize_token_embeddings regression: embeddings are resized but not propagated to model-specific input/output embedding modules." }, { - "left": "issue:43404", - "right": "issue:43454", + "left": "issue:44164", + "right": "issue:44291", "accept": false, - "reason": "Both are weight-tying problems, but in different model classes and likely different code paths." + "reason": "Both concern loading/saving, but one is extra_state handling and the other is init_empty_weights argument compatibility; different fixes." }, { - "left": "issue:44164", - "right": "issue:44291", + "left": "issue:43404", + "right": "issue:43454", "accept": false, - "reason": "save/from_pretrained extra_state handling and init_empty_weights argument compatibility are separate serialization/init regressions." + "reason": "Both involve tied lm_head weights, but they are different model families and different manifestations." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Qwen3 auto device map CUDA assert is unrelated to AudioFlamingo3 batched inference token/embedding leakage." + "reason": "Different Qwen3 device-map crash vs AudioFlamingo batch leakage; not the same underlying issue." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Tokenizer vocab-size mismatch for one model is not the same as the other model\u2019s loading failure." + "reason": "Both are vocab/tokenizer-related, but one is token IDs outside vocab size and the other is model loading failure; not a duplicate." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "SigLIP2 docs/example errors and a generic GLM5 issue are unrelated." + "reason": "SigLIP2 documentation/example errors are unrelated to GLM5 loading." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Torch 2.10 incompatibility is separate from meta-device loading regressions." + "reason": "Both involve compatibility regressions, but one is Torch version incompatibility and the other is meta-device loading breakage." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing with PEFT LoRA and T5 apex.FusedRMSNorm memory leak are unrelated." + "reason": "Gradient checkpointing with PEFT LoRA is unrelated to a T5 apex RMSNorm memory leak." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Qwen3.5 multimodal text_config unwrapping are different model/code-path bugs." + "reason": "Gemma4 FSDP2 shared_kv_states KeyError is unrelated to AutoModelForCausalLM.from_config text_config unwrapping." }, { - "left": "issue:43873", - "right": "issue:44387", + "left": "issue:43335", + "right": "issue:45663", "accept": false, - "reason": "Quantized offloading behavior and int4 reserved-memory/OOM are related only at a high level, not the same underlying bug." + "reason": "Different model bugs: SwitchTransformers sparse-layer construction vs Gemma4 attention/FSDP state handling." }, { - "left": "issue:43335", - "right": "issue:45663", + "left": "issue:43873", + "right": "issue:44387", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug is unrelated to Gemma4 attention state handling." + "reason": "Related to memory use, but offloading/quantization behavior is not the same as int4 reserved-memory regression." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Gemma4 FSDP2 KeyError and Qwen3.5 rotary embedding CPU-offload incompatibility are distinct issues." + "reason": "Gemma4 FSDP2 state rebuild bug is unrelated to Qwen3_5MoeTextRotaryEmbedding CPU offload incompatibility." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "init_empty_weights/_is_hf_initialized and Strix Halo mmap OOM are unrelated." + "reason": "init_empty_weights argument handling and disabling mmap on Strix Halo are different loading problems." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "AudioFlamingo3 batched inference leakage is not the same as Gemma4 PLE device mismatch." + "reason": "AudioFlamingo batch leakage and Gemma4 PLE device mismatch are unrelated model-specific bugs." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "GLM-4.7-Flash test loading and Qwen3.5 save_pretrained regression are separate model/version issues." + "reason": "Different models and failure modes: unit-test load failure vs incorrect saved checkpoint contents." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization formats, different models, and different load failures." + "reason": "Both are quantization loading issues, but for different models/quantization formats and likely different code paths." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Reserved-memory/OOM under quantization is not the same as disabling mmap on Strix Halo." + "reason": "Both touch memory behavior, but one is quantization OOM under Transformers 5.x and the other is mmap on Strix Halo; not the same bug." }, { "left": "issue:44743", "right": "issue:44960", "accept": false, - "reason": "Qwen3.5 recurrent-state reset bug and GLM5 are unrelated." + "reason": "Qwen modular recurrent-state reset bug is unrelated to GLM5 model-loading." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading regressions and Strix Halo mmap OOM are unrelated." + "reason": "Meta-device loading regressions and mmap-on-Strix-Halo OOM are different issues." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Save_pretrained permission propagation and wrong custom-module reload are both custom-model save/load topics, but not the same bug." + "reason": "Both involve save_pretrained/from_pretrained custom-model workflows, but one is file permission propagation and the other is loading the wrong custom module." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "These describe the same resize_token_embeddings propagation regression; 45276 is a Gemma4-specific manifestation and 45292 is the broader version." + "reason": "Same resize_token_embeddings regression; both report that resizing does not update output embeddings." }, { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Gradient checkpointing in eval mode is unrelated to Qwen3.5 position_ids causing SDPA fallback and NaNs." + "reason": "Gradient checkpointing in eval mode is unrelated to the Qwen3.5 3D position_ids / SDPA fallback issue." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "Serialization extra_state handling is unrelated to mmap-driven OOM behavior." + "reason": "save/from_pretrained extra_state handling is unrelated to mmap/OOM behavior." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "PEFT LoRA gradient checkpointing ineffectiveness and eval-mode checkpointing restriction are different gradient-checkpointing bugs." + "reason": "Gradient checkpointing with PEFT LoRA is a different bug from checkpointing being disallowed in eval mode." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays mostly conservative. The only accepted soft duplicate, issue:45276 -> issue:45292, is supported by the titles and the high similarity signal. The other pair verdicts are appropriately negative and do not overclaim duplicate status." + "feedback": "The summary is grounded in the packet and stays conservative. The only duplicate-like cluster with strong title-level overlap is the resize_token_embeddings group (45276/45292/45335), and the canonical issue choice is plausible from the broader title scope. The other soft-edge rejections are appropriately cautious and do not overstate shared root causes." }, "evaluator_used": true, "retried": false, @@ -2304,8 +2169,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -2364,8 +2229,8 @@ "issue:45588|issue:45663", "issue:45292|issue:45335", "issue:43645|issue:44164", - "issue:44402|issue:44451", "issue:43257|issue:44863", + "issue:44402|issue:44451", "issue:43828|issue:43957", "issue:43425|issue:43828", "issue:44589|issue:44756", @@ -2388,154 +2253,154 @@ "issue:44589|issue:45084", "issue:44164|issue:45759", "issue:36032|issue:43452", - "issue:44671|issue:44991", "issue:43596|issue:44683", + "issue:44671|issue:44991", "issue:32090|issue:39404", "issue:43856|issue:44756", "issue:44991|issue:45081" ], - "prepared_review_unit_hash": "f55f477ec51b0b0fe978b50169e92ad114689b461371d8cca72340eefc0850a9", + "prepared_review_unit_hash": "2e4cef05bb96ac5f430ade1e096c3acbf1de25e463daa8cdc92db5d8b1056216", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 38055, - "estimated_input_tokens": 9514, - "estimated_eval_tokens": 19284 + "serialized_chars": 38057, + "estimated_input_tokens": 9515, + "estimated_eval_tokens": 19286 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of Transformers 5.x regressions, especially model/tokenizer loading issues, but the soft pairs are generally different bugs affecting different code paths or model families. Only the two `resize_token_embeddings` reports look close enough to be treated as the same underlying issue.", - "confidence": 0.84, - "canonical_issue_reason": "Issue 43611 is the broadest representative of the cluster\u2019s main theme: a Transformers 5.0 loading regression. It is generic, model-agnostic, and best matches the cluster\u2019s repeated loading-breakage pattern.", + "summary": "This cluster is mostly heterogeneous Transformers v5 regression reports, with one clear duplicate pair around `resize_token_embeddings` behavior. Most other pairs only share a broad area (model loading, tokenizers, multimodal configs, docs) and should stay separate.", + "confidence": 0.43, + "canonical_issue_reason": "`issue:43611` is the broadest representative of the main theme in this cluster: Transformers v5 breaking model loading / initialization paths. It is not a perfect fit for every item, but it best captures the dominant regression pattern here.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43611 is the most suitable global representative because it is the most general and central loading regression among many heterogeneous model-specific failures.", + "best_issue_reason": "`issue:43611` is the best single issue to anchor the cluster because it describes a general v5 loading regression rather than a narrow model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Both are v5-era breakages, but they affect different models and different mechanisms: Qwen video input regression vs granite_speech strict config loading." + "reason": "Both are v5-era multimodal loading regressions, but they hit different models and different failure modes; not the same bug." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model initialization in notebooks is unrelated to incorrect `resume_from_checkpoint` max_steps calculation." + "reason": "Custom model initialization in notebooks is unrelated to `resume_from_checkpoint` / `max_steps` calculation." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Different failures in different subsystems: flash-attention handling of `s_aux=None` vs Gemma4 FSDP2 `shared_kv_states` rebuild behavior." + "reason": "Different code paths and failures: `s_aux=None` crash in flash attention vs Gemma4 FSDP2 `shared_kv_states` rebuild issue." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "These both describe the same underlying `resize_token_embeddings` propagation bug, with 45335 being a model-specific instance of the broader output-embedding mismatch." + "reason": "Same underlying `resize_token_embeddings` propagation bug: output/decoder embeddings are not updated consistently." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Jupyter custom-model initialization problems are not the same as `save/from_pretrained` extra_state handling." + "reason": "Both touch model lifecycle code, but notebook custom-model init and `extra_state` handling are distinct problems." }, { - "left": "issue:44402", - "right": "issue:44451", + "left": "issue:43257", + "right": "issue:44863", "accept": false, - "reason": "Tokenizer vocab-size mismatch for one repo is a different issue from a general inability to load another tokenizer in the latest version." + "reason": "Different models and different loading failures; not a shared concrete bug." }, { - "left": "issue:43257", - "right": "issue:44863", + "left": "issue:44402", + "right": "issue:44451", "accept": false, - "reason": "Different model families and different loading failures; no evidence they share the same root cause." + "reason": "One is a tokenizer vocab-size mismatch, the other is a general model load failure for a different checkpoint." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "An autocast dtype mismatch is not the same bug as meta-device loading breakage." + "reason": "Autocast dtype mismatch and meta-device loading breakage are different failure modes." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch version incompatibility is too broad and unrelated to the specific Phi MoE autocast dtype mismatch." + "reason": "Torch version incompatibility is not the same bug as the Phi MoE autocast dtype mismatch." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup failure and Strix Halo mmap OOM are unrelated runtime issues." + "reason": "Float8 storage lookup failure and mmap OOM on Strix Halo are unrelated." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 device-map mismatch are different code paths and symptoms." + "reason": "Compiled flex attention failure and Gemma4 device mismatch are different code paths." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper `return_language` pipeline regression is unrelated to tokenization changes." + "reason": "Whisper pipeline `return_language` regression is unrelated to the broad tokenization-change report." }, { "left": "issue:43931", "right": "issue:44863", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and NemotronH checkpoint loading failure are separate model-specific problems." + "reason": "Both are model-loading issues, but they involve different model families and different root causes." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "An isolated reproduction of another issue is not the same as the Qwen3.5 PagedAttentionCache crash." + "reason": "An isolated reproduction of a prior issue is not the same bug as the Qwen3.5 paged-attention crash." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "Torch 2.10 incompatibility is not the same as Gemma4 PLE device mismatch under `device_map=\"auto\"`." + "reason": "Torch 2.10 incompatibility and Gemma4 PLE device mismatch are unrelated." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "Generic model-loading regression and notebook custom-model initialization failure are different breakages." + "reason": "Both involve v5 breaking custom-model flows, but one is generic base-model loading and the other is notebook-specific initialization." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "Torch version incompatibility is unrelated to Mistral-3 image preprocessor/model dtype mismatch." + "reason": "Torch compatibility and Mistral-3 dtype mismatch are different issues." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and template-node compilation error are distinct failures." + "reason": "PagedAttention cache crash and template compilation error are different failures." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "These are both documentation-related, but they concern different deprecated pipeline APIs and are not the same change." + "reason": "Both are docs problems, but they cover different deprecated/removed pipeline task mentions." }, { "left": "issue:39404", @@ -2547,55 +2412,91 @@ "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "`torch.export` failures with `torch_compilable_check` and Perceiver resize/interpolation failure are different bugs." + "reason": "torch.export / `torch_compilable_check` is a different code path from Perceiver interpolation failure." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Qwen video-input regression and a generic mask-shape IndexError are not the same issue." + "reason": "Video input regression and mask-shape IndexError do not share the same underlying bug." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "Wrong SigLIP2 outputs are unrelated to `ModelOutput` key assignment when a field was previously `None`." + "reason": "Siglip2 bad outputs and `ModelOutput` key handling are unrelated." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Both are Qwen3-VL-related, but one is a broader loading regression and the other is a specific weight-shape mismatch; not enough evidence they are the same bug." + "reason": "Both mention Qwen3-VL loading, but one is MoE conversion and the other is checkpoint shape mismatch; not enough evidence of one concrete bug." }, { "left": "issue:39404", "right": "issue:44671", "accept": false, - "reason": "Whisper pipeline regression and CamemBERT masked-LM prediction issues are unrelated." + "reason": "Whisper return_language regression and CamemBERT masked LM behavior are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch incompatibility and Qwen3.5 input-embedding unpacking are different problems." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage lookup and template compilation failure are unrelated." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "`extra_state` save/load handling and `from_config` text_config unwrapping are different issues." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5 tokenizer special-token conflict and GGUF file handling are unrelated tokenizer/model-loading bugs." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "DeepSpeed ZeRO3 init IndexError and compiled flex attention failure are different code paths." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT masked-LM predictions and tokenizer loading for est-roberta are not the same bug." }, { "left": "issue:32090", "right": "issue:39404", "accept": false, - "reason": "Trainer GPU broadcast type error and Whisper pipeline `return_language` regression have no shared code path." + "reason": "Trainer GPU broadcast `NoneType` error and Whisper pipeline regression are unrelated." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Qwen3 MoE training memory usage and Strix Halo mmap OOM are different performance/memory issues." + "reason": "Qwen3 MoE memory usage and Strix Halo mmap OOM are different performance/resource issues." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Tokenizer loading regression for est-roberta and Mistral regex patch crashing on missing `backend_tokenizer` are different tokenizer bugs." + "reason": "Tokenizer load failure and Mistral regex patch crash are related only broadly by tokenizer code, not the same concrete defect." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary stays grounded in the packet: 43611 is a defensible broad representative, and the overall characterization as mostly Transformers 5.x regression/loading issues matches the listed issue titles. The soft-edge verdicts are conservative, with the one accepted pair (45292/45335) plausibly describing the same resize_token_embeddings bug, while the rest are reasonably treated as distinct failure modes or model families." + "feedback": "The cluster summary is grounded in the packet and stays appropriately conservative. The broad characterization as mostly heterogeneous v5 regression reports is supported, and the single accepted duplicate pair (45292/45335) looks plausibly like the same resize_token_embeddings bug. The other soft-edge rejections do not overclaim sameness." }, "evaluator_used": true, "retried": false, @@ -2611,8 +2512,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -2672,8 +2573,8 @@ "issue:44488|issue:45081", "issue:44661|issue:44683", "issue:40990|issue:43421", - "issue:43611|issue:43644", "issue:44488|issue:44987", + "issue:43611|issue:43644", "issue:44206|issue:44220", "issue:43716|issue:45084", "issue:43883|issue:44488", @@ -2699,220 +2600,238 @@ "issue:43322|issue:43540", "issue:43645|issue:44534" ], - "prepared_review_unit_hash": "5c335803df76a9527f5824145b65269649e424db6c3e610803322a6191511a91", + "prepared_review_unit_hash": "a48c46421f835c57885b7444fe17437c40aa4cc5313d830b46738e394145001f", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37331, - "estimated_input_tokens": 9333, - "estimated_eval_tokens": 18922 + "serialized_chars": 37328, + "estimated_input_tokens": 9332, + "estimated_eval_tokens": 18920 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a grab-bag of unrelated issue reports, many in the same broad Transformers v5 regression era but not the same concrete bug. I would not merge the soft pairs under duplicate triage based on titles alone.", - "confidence": 0.71, - "canonical_issue_reason": "Issue 43646 is the most umbrella-like representative here: it describes a broad custom-model initialization regression in v5, which is closer to a cluster headline than the narrower model- or feature-specific reports.", + "summary": "This cluster is mostly a mix of unrelated issue reports, with only a couple of clear duplicate pockets around Transformers v5 regressions: non-persistent buffer serialization and custom model initialization. I would keep the non-persistent-buffer bug as the main canonical issue.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 43644 is the most concrete representative of the duplicated v5 buffer-regression cluster, and its wording is essentially repeated by 44534.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43646 is the best single issue to represent the overall mixed set because its title is broad and generic enough to cover a family of adjacent v5 initialization problems.", + "best_issue_reason": "Issue 43644 is the best single issue to anchor the cluster because it is specific, directly duplicated, and captures the most obvious repeated failure mode.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both are device/offload-related, but they affect different model code paths and different failure modes." + "reason": "Different models and symptoms: XLNet CPU device-placement bug vs Qwen3.5 MoE CPU-offload incompatibility." }, { "left": "issue:44164", "right": "issue:44479", "accept": false, - "reason": "Serialization/save-load extra_state handling is unrelated to the video-input regression." + "reason": "Unrelated areas: save/from_pretrained extra_state handling vs video-input regression." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Apple Silicon allocator warmup and _is_hf_initialized are different regressions." + "reason": "Apple Silicon allocator TypeError and audio chat-template sampling-rate bug are unrelated." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Different tokenizer-loading failures for different models; no shared concrete bug is evident." + "reason": "Both are tokenizer-loading complaints, but for different models and different failures; not the same concrete bug." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Tokenizer mapping bug and flex_attention compilation bug are unrelated." + "reason": "Tokenizer mapping helper failure and compiled flex_attention regression are unrelated." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity on a specific model is not the same as runtime post-processor update behavior." + "reason": "Perplexity regression and runtime special-token/post-processor update bug do not share the same code-path." }, { - "left": "issue:43611", - "right": "issue:43644", + "left": "issue:44488", + "right": "issue:44987", "accept": false, - "reason": "Both are v5 regressions, but one is model loading with base_model_prefix and the other is buffer initialization junk." + "reason": "Both mention loading failures, but they involve different model families and different loader paths." }, { - "left": "issue:44488", - "right": "issue:44987", + "left": "issue:43611", + "right": "issue:43644", "accept": false, - "reason": "Both mention loading failures, but they target different models and likely different code paths." + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer serialization." }, { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "These both point at audio feature extraction, but the titles are too vague to confirm the same underlying defect." + "reason": "Same broad audio subsystem, but the titles point to different feature-extraction failures." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Image-preprocessor dtype mismatch and template compilation failure are unrelated." + "reason": "Mistral-3 image dtype mismatch and template-compilation error are unrelated." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Missing all_tied_weights_keys and model-loading failure for a specific checkpoint are different issues." + "reason": "Different model-loading breakages with no clear shared bug." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "trust_remote_code config field loss is unrelated to PagedAttentionCache group-type handling." + "reason": "trust_remote_code config field loss and Qwen3.5 cache/group-type crash are unrelated." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "Trainer token counter update and multi-image IterableDataset finetuning are different problems." + "reason": "Trainer token accounting and multimodal iterable-dataset failure are unrelated." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both involve Whisper and special tokens, but one is batch_decode behavior and the other is empty-transcription generation." + "reason": "Both are Whisper-related, but one is batch_decode skip_special_tokens and the other is empty-transcription generation after align_special_tokens." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Tokenizer method conflict and FP8 config failure are unrelated." + "reason": "Tokenizer method-name conflict and Qwen3Moe FP8 failure are unrelated." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Both are v5 regressions, but one is non-persistent buffer serialization and the other is custom model initialization." + "reason": "Non-persistent buffer junk and custom model initialization are adjacent v5 regressions, but not the same concrete bug." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Different model-loading failures with no shared bug identity." + "reason": "Both are loading regressions, but for different models and different failure modes." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "Both involve GGUF loading, but one is a generic gguf_file path issue and the other is a specific Qwen2.5-GGUF failure." + "reason": "Both involve GGUF loading, but one is a generic from_pretrained breakage and the other is a specific Qwen2.5-GGUF v5 failure; not enough to merge strictly." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Torch dynamo grouped_mm tracing and template-node compilation errors are unrelated." + "reason": "Torch dynamo grouped_mm tracing failure and template compilation error are unrelated." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "FP8 config failure and linear_attention group-type crash are different model/backend issues." + "reason": "Different Qwen3-family bugs: FP8 config failure vs linear_attention cache crash." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "One is output_hidden_states behavior, the other is rope index scaling for Qwen2.5-VL." + "reason": "Hidden-states bug in Qwen3.5 and Qwen2.5-VL rope-index scaling bug are different code paths." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "Dequantized model offload failure and GGUF loading failure are not the same bug." + "reason": "Dequantized-model device_map offload failure and Qwen2.5-GGUF loading regression are not the same bug." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "Tokenizer special-token method conflict is unrelated to notebook custom-model initialization." + "reason": "Tokenizer special-method conflict and custom model initialization regression are unrelated." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Different model/tokenizer load failures; not enough to treat as one defect." + "reason": "Both are version-gated loading failures, but they affect different artifacts and loader paths." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Missing tied-weights metadata and tokenizer loading failure are unrelated." + "reason": "Different model-loading issues with no shared concrete root cause." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Both mention lm_head.weight, but one is untied weights in a specific model and the other is device-dependent serialization behavior." + "reason": "Mistral3 lm_head tying bug and inconsistent tied-weight serialization are related only at a very broad theme level." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Template compilation and GraniteMoEHybrid invalid-method calls are different failures." + "reason": "Template compilation error and GraniteMoEHybrid invalid-method call are unrelated." }, { "left": "issue:43262", "right": "issue:43377", "accept": false, - "reason": "Audio chat-template sampling-rate handling and MIMI padding-mask behavior are different issues." + "reason": "Audio chat-template sampling-rate bug and MIMI padding-mask bug are different modalities and code paths." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Both are vision-model mismatches, but the models and failure modes differ." + "reason": "SigLIP2 nonsensical outputs and OmDet-Turbo image-size mismatch are unrelated." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Custom model initialization and non-persistent buffer junk are adjacent v5 regressions, but not the same concrete bug." + "reason": "Custom model initialization regression and non-persistent buffer serialization are different failures." }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Llava Next load-time segfault and Qwen3OmniMoe video-processing ValueError are unrelated." + "reason": "Llava Next segfault and Qwen3OmniMoe video ValueError are different multimodal loading/processing bugs." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Notebook custom-model init breakage and non-persistent buffer serialization are not the same bug." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "These are the same custom-model initialization regression, with the notebook case just being a more specific manifestation." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Same v5 non-persistent-buffer junk regression; the titles are essentially the same bug." } ] }, @@ -2931,8 +2850,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -2989,8 +2908,8 @@ "issue:43377|issue:43819", "issue:42915|issue:45084", "issue:44265|issue:44610", - "issue:45216|issue:45357", "issue:44451|issue:45081", + "issue:45216|issue:45357", "issue:44987|issue:45081", "issue:44220|issue:44683", "issue:43404|issue:45127", @@ -3007,30 +2926,30 @@ "issue:36032|issue:41720", "issue:43425|issue:44403", "issue:44079|issue:45663", - "issue:41553|issue:43582", "issue:40990|issue:44811", + "issue:41553|issue:43582", "issue:43632|issue:44534", "issue:42175|issue:45542", "issue:43421|issue:44568", "issue:43818|issue:45357", "issue:43792|issue:44589", - "issue:44683|issue:45020", - "issue:44403|issue:44991" + "issue:44403|issue:44991", + "issue:44683|issue:45020" ], - "prepared_review_unit_hash": "53f4f80e14a7c75ccc9bf50313cf987110f456b823cf96ab691270b954db2c29", + "prepared_review_unit_hash": "ab691ecabb728ea9eab336aea30a7cc55eb46fb8abc104b3459cfe742584975c", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36438, + "serialized_chars": 36439, "estimated_input_tokens": 9110, "estimated_eval_tokens": 18476 }, @@ -3039,208 +2958,205 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose collection of unrelated Transformers bugs. The only likely true duplicate pair is the Qwen3.5 save_pretrained regression (45216/45357); everything else looks like separate issues affecting different models, code paths, or subsystems.", - "confidence": 0.64, - "canonical_issue_reason": "issue:45216 is the broader Qwen3.5 save_pretrained regression report and best matches the likely duplicate follow-up in 45357.", + "summary": "This cluster is mostly a loose collection of unrelated Transformers regressions, with a noticeable concentration around v5 model/tokenizer loading behavior. None of the soft pairs is close enough to be treated as a true duplicate or merge-worthy PR-equivalent.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43632 is the most central representative of the dominant theme: a core Transformers v5 load-path regression involving internal initialization state, which plausibly underlies several later loading failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the broadest open issue in the set and serves as the strongest general representative of the recent-model breakage theme, even though most items are not duplicates of it.", + "best_issue_reason": "Issue 43632 is the clearest cluster anchor because it describes a broad v5 regression in model loading, whereas most other issues are narrower, model-specific, or only loosely related.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Different problems: generic loading noise vs a Mistral tokenizer regex-patching crash." + "reason": "Different bugs: noisy load-time logging vs a tokenizer regex crash." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "MIMI padding-mask batching bug is unrelated to DAC.from_latents/STE mismatch." + "reason": "Unrelated code paths: MIMI padding-mask batching bug vs DAC latent reconstruction mismatch." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "Qwen3Moe FP8 config failure and template-node compilation error are separate code paths." + "reason": "Different failures: Qwen3Moe FP8 config handling vs template-node compilation error." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "Torch.export with torch_compilable_check is unrelated to OmDet-Turbo's processor/model input-size mismatch." - }, - { - "left": "issue:45216", - "right": "issue:45357", - "accept": true, - "reason": "Both describe the same Qwen3.5 save_pretrained regression; 45357 is a more specific follow-up about incorrect visual encoder keys." + "reason": "Different problems in different subsystems: torch.export/compilable checks vs image-size processor mismatch." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "ScandiBERT tokenizer loading failure is not the same as the Mistral regex patch crash." + "reason": "Both are load failures, but for different models and different root causes." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": false, + "reason": "Same model family, but one is a saved-checkpoint regression and the other is incorrect visual-encoder key saving; not the same concrete bug." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Physical-intelligence/fast loading failure and Mistral tokenizer patching are different bugs." + "reason": "Different model-loading failures with no shared root cause." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Feature-extraction fbank issue is unrelated to compiled flex_attention compatibility on newer torch." + "reason": "Different subsystems: fbank feature extraction vs compiled flex_attention on newer torch." }, { "left": "issue:43404", "right": "issue:45127", "accept": false, - "reason": "Both touch tied weights, but one is a Mistral3 load-time bug and the other is LoRA merge with extended vocabulary." + "reason": "Both touch weight tying, but one is a specific missing lm_head tie and the other is LoRA merge collapse after vocab extension." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "_is_hf_initialized flag breakage and mmap OOM on Strix Halo are unrelated." + "reason": "Unrelated: internal v5 initialization flag break vs mmap OOM on Strix Halo." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "Both involve model initialization, but one is FineGrainedFP8Config and the other is an unexpected init_empty_weights argument." + "reason": "Different root causes: Qwen3Moe FP8 config failure vs init_empty_weights argument compatibility." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "AddedToken special-argument conflict is unrelated to loading physical-intelligence/fast." + "reason": "Tokenizer AddedToken argument error is unrelated to the physical-intelligence/fast load regression." }, { "left": "issue:44479", "right": "issue:45325", "accept": false, - "reason": "Video input regression across Qwen VL variants is not the same as still-image RoPE scaling." + "reason": "Both are Qwen VL regressions, but one is broad video-input behavior and the other is a specific still-image rope-index scaling bug." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "T5Tokenizer add_special_tokens conflict and MobileLLM key error are different tokenizer/model-loading failures." + "reason": "Different failures: T5 tokenizer method conflict vs MobileLLM load key error." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "PaddleOCR-VL load failure in vLLM and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "Different model-specific failures: PaddleOCR-VL loading in vLLM vs Qwen3-MoE sliding_window behavior." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "Multi-image IterableDataset training failure is distinct from Qwen2.5-VL temporal position scaling." + "reason": "Different bugs: iterable dataset with multiple images vs temporal position scaling for still images." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "FP8 config failure is not the same as the missing Float8 storage TypeError." + "reason": "FP8 config failure and missing Float8 storage are related only at a very high level." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "base_model_prefix loading breakage is unrelated to junk-filled non-persistent buffers." + "reason": "Different Transformers v5 regressions: base_model_prefix loading vs junk-filled non-persistent buffers." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Qwen3-MoE sliding_window issue and granite_speech strict config loading are separate bugs." + "reason": "Both mention config/loading issues, but they affect different models and different mechanisms." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "Tokenizer attribute conflict is unrelated to the Qwen3 auto device mapping cudaErrorAssert." + "reason": "Tokenizer-loading conflict vs CUDA assert during auto device mapping are unrelated." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch 2.10 incompatibility is not the same as generic transformer loading noise." + "reason": "Torch version incompatibility is unrelated to a noisy load-message issue." }, { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "ModelOutput key assignment and Gemma4 shared_kv_states/FSDP2 are unrelated." + "reason": "ModelOutput key assignment bug and Gemma4 FSDP2 shared_kv_states bug are different code paths." }, { - "left": "issue:41553", - "right": "issue:43582", + "left": "issue:40990", + "right": "issue:44811", "accept": false, - "reason": "Bad AutoTokenizer error for Voxtral and Apple Silicon caching_allocator_warmup TypeError do not match." + "reason": "Perplexity regression on GPT-OSS is unrelated to Whisper batch_decode special-token handling." }, { - "left": "issue:40990", - "right": "issue:44811", + "left": "issue:41553", + "right": "issue:43582", "accept": false, - "reason": "High perplexity on GPT-OSS and Whisper batch_decode skip_special_tokens are different issues." + "reason": "Bad tokenizer error messaging and Apple Silicon allocator TypeError are unrelated." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "Transformers v5 _is_hf_initialized breakage is separate from non-persistent buffer junk values." + "reason": "Different v5 regressions: initialization flag handling vs buffer initialization contents." }, { "left": "issue:42175", "right": "issue:45542", "accept": false, - "reason": "Missing TensorFlow in the torch extra and tensorboard-only undefined tf backend error are related but not the same bug." + "reason": "Both concern TensorFlow packaging, but one is missing TF in the torch extra and the other is tensorboard-only causing an undefined backend error." }, { "left": "issue:43421", "right": "issue:44568", "accept": false, - "reason": "Runtime post-processor refresh behavior is different from add_special_tokens not adding BOS/EOS for mdeberta-v3-base." + "reason": "Both involve special tokens, but one is runtime post-processor syncing and the other is add_special_tokens not inserting BOS/EOS." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Video-LLaVA weight-sharing/temporal-attention bug is unrelated to Qwen3.5 save_pretrained key issues." + "reason": "Different bugs: Video-LLaVA tower architecture/weights vs Qwen3.5 save_pretrained key serialization." }, { "left": "issue:43792", "right": "issue:44589", "accept": false, - "reason": "Whisper-large-v2 runtime failure and missing Float8 storage are different problems." + "reason": "Whisper runtime failure and missing Float8 storage are unrelated." }, { - "left": "issue:44683", - "right": "issue:45020", + "left": "issue:44403", + "right": "issue:44991", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 is not the same as remote_code loading breakage." + "reason": "Generic load noise and a tokenizer load failure for a specific model are not the same bug." }, { - "left": "issue:44403", - "right": "issue:44991", + "left": "issue:44683", + "right": "issue:45020", "accept": false, - "reason": "Generic loading noise is distinct from the EMBEDDIA/est-roberta tokenizer load failure." + "reason": "Compiled flex_attention failure on torch>=2.9 is unrelated to remote_code loading regressions." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft-edge pair, 45216/45357, is presented as a likely duplicate/follow-up rather than a hard certainty, and the rest are reasonably rejected as distinct bugs. No obvious overstatement of 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 }, @@ -3324,8 +3240,8 @@ "issue:41720|issue:45081", "issue:43531|issue:44291", "issue:44265|issue:45072", - "issue:45325|issue:45381", "issue:43632|issue:45072", + "issue:45325|issue:45381", "issue:43994|issue:44898", "issue:43611|issue:43646", "issue:42175|issue:43296", @@ -3338,20 +3254,20 @@ "issue:43701|issue:45325", "issue:44479|issue:45823" ], - "prepared_review_unit_hash": "b29c25ef3c30815591f7c88a1fcf130494eb8a923d254cdd6be892028d4c175f", + "prepared_review_unit_hash": "46d4cd011c4b00cffd0bd48c91c8946f1cb9ec6a129c8d57f5e79d412a00318d", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36750, + "serialized_chars": 36752, "estimated_input_tokens": 9188, "estimated_eval_tokens": 18632 }, @@ -3360,201 +3276,201 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous. The only likely duplicate subcluster is the Qwen2.5-VL position-id handling regression around `get_rope_index` / `vision_position_ids` in v5.3.0; the rest look like separate model-specific or subsystem-specific bugs.", - "confidence": 0.47, - "canonical_issue_reason": "45325 is the clearest anchor because it names the specific Qwen2.5-VL `get_rope_index` / temporal position-id path in transformers 5.3.0, which is the most concrete shared mechanism visible in this packet.", + "summary": "Mostly distinct Transformers regressions across different models/subsystems. A small number of pairs look close enough to be duplicates, especially the v5 model-initialization/loading reports and the Qwen2.5-VL position-id regressions.", + "confidence": 0.64, + "canonical_issue_reason": "issue:43611 is the broadest and clearest representative of the v5 model-loading / initialization regression class, with a generic scope that can absorb closely related reports.", "canonical_pr_reason": null, - "best_issue_reason": "45325 is the best representative issue for this cluster because it is the most specific report of the likely Qwen2.5-VL position-id regression; 45381 looks like a related modality-specific symptom rather than a broader anchor.", + "best_issue_reason": "issue:43611 is the best anchor issue for this cluster because it describes a concrete, widely applicable Transformers v5 breakage rather than a narrow model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44462", "right": "issue:45812", "accept": false, - "reason": "Same tokenizer area, but `tokenizer.json` loading and Granite token-ID regression are different failures." + "reason": "Both are tokenizer regressions, but the symptoms differ too much: one is repository tokenizer.json lookup, the other is Granite-only token-id corruption." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Both mention Qwen/VL, but one is multi-GPU training and the other is multi-image IterableDataset finetuning; not the same bug." + "reason": "Different multimodal training failures: multi-GPU training vs iterable-dataset multi-image prompting." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Both are sequence-classification related, but the titles point to different failure modes and likely different code paths." + "reason": "Both mention degenerate training, but one is an attention backend bug and the other is a classification-loss setup bug." }, { "left": "issue:43317", "right": "issue:43856", "accept": false, - "reason": "One is `device_map=auto` loading/offload failure; the other is memory usage during Qwen3 MoE training." + "reason": "Different problems: device_map/offload loading failure vs MoE training memory inefficiency." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "GGUF loading failure and `CLIPTokenizer` `model_max_length` behavior are unrelated." + "reason": "Unrelated tokenization/model-loading issue versus a CLIPTokenizer max-length constant bug." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing in eval mode and missing `supports_gradient_checkpointing` are different issues." + "reason": "One is an eval-mode checkpointing restriction; the other is a missing model capability flag on trust_remote_code." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Progressive generation with `inputs_embeds`/`past_key_values` is unrelated to multi-image finetuning." + "reason": "Different generation/training scenarios and different failure modes." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "ModelOutput key assignment and non-persistent buffer serialization are distinct regressions." + "reason": "ModelOutput key assignment bug is unrelated to non-persistent buffer serialization junk." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Different loading errors: GGUF/transformers v5 vs missing `pad_token_id` on `Llama4Config`." + "reason": "Both are loading-related, but one is GGUF/v5 loading and the other is a missing config attribute on Llama4." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "`init_empty_weights` argument handling and Float8 storage lookup failures are separate problems." + "reason": "Different loading-time failures: unexpected init argument vs missing Float8 storage class." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "MIMI batching/padding-mask behavior is unrelated to Gemma3 sliding-window NaN embeddings." + "reason": "Both involve batching/padding, but they affect different models and different code paths." }, { "left": "issue:43749", "right": "issue:44079", "accept": false, - "reason": "FSDP CPU RAM-efficient loading and ModelOutput key assignment do not share the same underlying bug." + "reason": "FSDP CPU RAM efficient loading is unrelated to ModelOutput key handling." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL issues, but one is multi-image finetuning and the other is position-id handling." + "reason": "Both concern Qwen2.5-VL/video-image inputs, but the reported bugs are different enough to keep separate." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Non-persistent buffer junk in v5 is unrelated to ModelOutput key assignment." + "reason": "Non-persistent buffer filling is not the same issue as ModelOutput key assignment." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Qwen3 auto device mapping crash and Mistral tokenizer backend error are unrelated." + "reason": "Completely different failures: CUDA assert under device mapping vs tokenizer regex patch crash." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Qwen3-MoE `sliding_window` behavior and `init_empty_weights` argument errors are different defects." + "reason": "Different regressions in Qwen3-MoE sliding-window behavior versus init_empty_weights argument handling." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "`torch.export`/`torch_compilable_check` failure is unrelated to bfloat16 dtype mismatches." - }, - { - "left": "issue:45325", - "right": "issue:45381", - "accept": true, - "reason": "Both point to Qwen2.5-VL position-id handling in v5.3.0; 45381 looks like a video-input manifestation of the same underlying regression." + "reason": "torch.export compatibility and dtype mismatches are different concrete bugs." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "`_is_hf_initialized` breakage and bfloat16 dtype mismatch are different Transformers v5 regressions." + "reason": "_is_hf_initialized flag breakage is unrelated to bfloat16 dtype mismatches." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": true, + "reason": "Both point to the same Qwen2.5-VL 5.3.0 position-id / rope-index generation regression, just surfaced through still-image vs video paths." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "SigLIP2 wrong outputs and Perceiver resize/interpolation failure are unrelated." + "reason": "Different vision-model bugs: SigLIP2 tokenization/output issue vs Perceiver resolution/interpolation failure." }, { "left": "issue:43611", "right": "issue:43646", - "accept": false, - "reason": "Both are v5 loading-related, but base-model-prefix loading and custom initialization are not clearly the same bug." + "accept": true, + "reason": "Both describe Transformers 5 breaking custom-model loading/initialization on the same code path, and could plausibly be fixed together." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Packaging/backends missing TensorFlow and PaddleOCR-VL loading failure are unrelated." + "reason": "Packaging/backend inclusion and model-loading failures are unrelated." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "AyaVision `lm_head` tying and device-dependent serialization inconsistency are related thematically, but not the same concrete issue." + "reason": "One is tied lm_head weights causing bad generation; the other is serialization inconsistency of lm_head.weight across devices." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Qwen2.5-VL position-id regression and Kosmos2.5 OCR index error are different model bugs." + "reason": "Position-id scaling bug and Kosmos2.5 long-OCR index error are different multimodal issues." }, { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Whisper runtime failure and meta-device loading breakage are unrelated." + "reason": "Different model-loading/runtime failures with no clear shared mechanism." }, { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "Qwen3Moe FP8 config failure and Qwen2.5-GGUF loading failure are different bugs." + "reason": "Qwen3MoE FP8 config failure is unrelated to Qwen2.5 GGUF loading under v5." }, { "left": "issue:43296", "right": "issue:45800", "accept": false, - "reason": "PaddleOCR-VL loading and torch/transformers version incompatibility are unrelated." + "reason": "PaddleOCR-VL loading failure and torch/transformers version incompatibility are different root causes." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Int4 CUDA reserved-memory OOM and Sam3Video OOM are different memory issues with different causes." + "reason": "Int4 reserved-memory regression and Sam3Video OOM are not the same bug." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL position-id handling." + "reason": "Checkpoint-resume key mismatch is unrelated to Qwen2.5-VL position-id generation." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Qwen video-input regression and Gemma4 device mismatch are different model-specific failures." + "reason": "Video-input regression in Qwen-family models is unrelated to Gemma4 device mismatch during forward." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster-level summary is broadly reasonable, but the representative/canonical issue rationale overstates what the packet supports. In particular, \"concrete, reproducible\" and \"the only item with a clearly plausible duplicate counterpart\" are not grounded here, since the packet only gives titles/body lengths and there are several other plausible soft-edge pairs. The 45325\u219445381 duplicate call may be plausible, but it should be phrased more tentatively rather than as the same code-path regression or broader bug without direct evidence." + "accept": true, + "feedback": "The cluster summary is grounded in the packet: issue:43611 is a reasonable broad anchor for the v5 model-loading/init regression class, and the two accepted soft edges are conservative enough based on their titles. The false soft-edge calls also stay appropriately narrow and do not overclaim duplicate status." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -3632,14 +3548,14 @@ "issue:44186|issue:44265", "issue:35707|issue:41720", "issue:43606|issue:43819", - "issue:43454|issue:43550", "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:43377|issue:44898", "issue:43454|issue:45072", + "issue:43377|issue:44898", "issue:43257|issue:45923", "issue:43526|issue:43883", "issue:40990|issue:42947", @@ -3653,14 +3569,14 @@ "issue:44008|issue:45823", "issue:43404|issue:45357" ], - "prepared_review_unit_hash": "21bd7e95c4155aa8d88207d83b472a371771b6c18c765cec9339101d057e8fd0", + "prepared_review_unit_hash": "051f45a1991f839b32b3d94781da8de31ce7612c59c3f65f90893c54ae915878", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, @@ -3675,198 +3591,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a set of mostly distinct bug reports spanning loading/offloading, tokenizer regressions, multimodal model mismatches, and model-specific forward errors. None of the soft pairs look like the same concrete bug/change, so I would not merge any of them.", - "confidence": 0.68, - "canonical_issue_reason": "issue:43873 is the broadest representative of the shared offloading/quantization theme and is open with active discussion, making it the best anchor for this loose cluster.", + "summary": "These items are mostly unrelated regressions spanning loading, tokenization, multimodal processing, and training/runtime bugs. None of the soft pairs look like true duplicates; the closest recurring theme is offloading/device_map/quantization loading behavior, where issue 43873 is the broadest representative.", + "confidence": 0.63, + "canonical_issue_reason": "Issue 43873 is the broadest and most representative of the recurring offloading/quantization-loading failures in this set, with active discussion and a general symptom description.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the most central issue for the overlapping loading/offloading regressions; it is broad enough to represent the cluster without being tied to one model.", + "best_issue_reason": "Issue 43873 is the best global representative because it captures the broadest underlying failure mode among the candidates and is currently open/actively relevant.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43881", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading failures, but they concern different models and likely different failure modes." + "reason": "Different failures in different models: glm-4v-9b load failure vs tokenizer/model loading for cjvt/sleng-bert." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU offload device mismatch is not the same bug as torch.export failing on torch_compilable_check." + "reason": "CPU offload device mismatch in Bark is unrelated to torch.export failures from torch_compilable_check." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "One is an init_empty_weights argument regression; the other is strict config handling for granite_speech." + "reason": "Unexpected _is_hf_initialized arg during init_empty_weights is a different bug from strict granite_speech config loading." }, { "left": "issue:44568", "right": "issue:45920", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different tokenization paths and different model families." + "reason": "One is add_special_tokens/BOS-EOS behavior for a specific tokenizer; the other is wrong token IDs across several models." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sample-rate defaulting is unrelated to PEFT adapter loading from local checkpoints." + "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech PEFT local-checkpoint loading are separate code paths." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "High perplexity on a benchmark is not the same as whisper generation failing on empty transcription." + "reason": "High perplexity on gpt-oss-20b is unrelated to Whisper empty-transcription failure after align_special_tokens." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "Bark voice_preset handling and progressive generation with inputs_embeds/past_key_values are separate code paths." + "reason": "Bark voice_preset not working is not the same as progressive generation with inputs_embeds and past_key_values." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Offloading/quantization behavior is unrelated to version-dependent tokenization changes." + "reason": "Offloading with quantization is unrelated to tokenizer changes across transformers versions." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer crash on NER/padding is different from export-time torch_compilable_check failures." + "reason": "LayoutLMv2Tokenizer padding/NER crash is a tokenizer bug, not a torch.export/torch_compilable_check issue." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Both involve generation/runtime issues, but they are distinct failures in different models and execution setups." + "reason": "Different runtime problems: generation with inputs_embeds/past_key_values versus Qwen3 device-map cudaErrorAssert." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "Device-mismatch on CPU offload is unrelated to DAC.from_latents missing STE behavior." + "reason": "Both mention model loading behavior, but Bark CPU-offload mismatch and DAC.from_latents/STE mismatch are distinct bugs." }, { - "left": "issue:43454", - "right": "issue:43550", + "left": "issue:41720", + "right": "issue:45823", "accept": false, - "reason": "Weight tying bugs and torch.compile SDPA failures are different concrete problems." + "reason": "Both involve device_map=auto, but they affect different models and failure modes (cuda assert vs forward device mismatch)." }, { - "left": "issue:41720", - "right": "issue:45823", + "left": "issue:43454", + "right": "issue:43550", "accept": false, - "reason": "Both are device_map-related, but they affect different models and different runtime failures." + "reason": "lm_head weight tying in AyaVision is unrelated to Bamba torch.compile SDPA failures." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "MIMI batch-padding behavior is unrelated to OmDet processor input-size mismatch." + "reason": "MIMI padding-mask inconsistency and OmDet-Turbo processor/model resolution mismatch are unrelated." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Both mention offload/quantization, but one is a dequantized-model load failure and the other is a broader offloading complaint; not clearly the same bug." + "reason": "Dequantized-model loading with gpu+cpu offload is related in theme but not the same concrete bug as offloading with quantization." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "Documentation/example mistakes for SigLIP2 are unrelated to Qwen2.5-VL iterable-dataset finetuning failures." + "reason": "SigLIP2 docs/example errors and Qwen2.5-VL iterable-dataset multi-image finetuning failures are different problems." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "A missing pad_token_idx config field is not the same as AutoModelForCausalLM failing to unwrap text_config." + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to not unwrapping text_config for multimodal Qwen configs." }, { - "left": "issue:43377", - "right": "issue:44898", + "left": "issue:43454", + "right": "issue:45072", "accept": false, - "reason": "Different image/model-size behavior in unrelated vision models; not the same underlying issue." + "reason": "AyaVision tied-weights bug is unrelated to dtype mismatches in bfloat16 inference." }, { - "left": "issue:43454", - "right": "issue:45072", + "left": "issue:43377", + "right": "issue:44898", "accept": false, - "reason": "lm_head weight tying and bfloat16 dtype mismatches in inference are different bugs." + "reason": "MIMI batched-vs-single output mismatch from padding masks is unrelated to Perceiver interpolation failure." }, { "left": "issue:43257", "right": "issue:45923", "accept": false, - "reason": "MoE weight conversion during accelerate+deepspeed loading is unrelated to a missing supports_gradient_checkpointing flag." + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is unrelated to a missing supports_gradient_checkpointing flag." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "reduce_labels returning the wrong shape and a missing all_tied_weights_keys attribute are unrelated." + "reason": "BEiT reduce_labels returning one label is unrelated to Molmo missing all_tied_weights_keys." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "Perplexity evaluation issues are unrelated to gradient checkpointing behavior with PEFT LoRA." + "reason": "Perplexity regression and gradient-checkpointing inefficacy are different training/evaluation issues." }, { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both are serialization/regression reports, but they target different multimodal save/load paths and different symptoms." + "reason": "lm_head serialization inconsistency by device is distinct from Qwen3.5 visual-encoder key serialization." }, { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "A specific model loading failure is not the same as a tokenizer-loading regression for another repository." + "reason": "Model loading failure for glm-4v-9b is unrelated to tokenizer loading for EMBEDDIA/est-roberta." }, { "left": "issue:43749", "right": "issue:44534", "accept": false, - "reason": "FSDP CPU RAM efficient loading and non-persistent buffer initialization are different bugs." + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk are separate regressions." }, { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "Weight tying for a multimodal model is unrelated to BitNet packed-weight unpacking during accelerate loading." + "reason": "AyaVision weight tying and BitNet packed-weight unpacking are different loading/serialization bugs." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Auto device-mapping cudaErrorAssert and Llava Next segfaults are different failure modes in different models." + "reason": "Qwen3 device-map cuda assert and Llava Next segfault are different model-specific failures." }, { "left": "issue:43296", "right": "issue:44877", "accept": false, - "reason": "vLLM loading a PaddleOCR-VL model and strict config loading for granite_speech are not the same problem." + "reason": "PaddleOCR-VL load failure in vLLM is unrelated to strict granite_speech config loading." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Both involve Qwen3.5 runtime behavior, but one is device-mapping failure and the other is flash-attention generation quality/behavior." + "reason": "Qwen3.5 flash-attention generation bug is not the same as Qwen3 auto device mapping cuda assert." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Variable-name collision in Gemma3n forward is unrelated to Gemma4 PLE device mismatch." + "reason": "Gemma 3n variable-name collision in forward is unrelated to Gemma4 PLE device mismatch." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "lm_head weight tying in Mistral3 and incorrect visual encoder keys in Qwen3.5 save_pretrained are different bugs." + "reason": "Mistral3 lm_head tying and Qwen3.5 visual encoder key serialization are separate issues." } ] }, @@ -3885,8 +3801,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -3962,8 +3878,8 @@ "issue:43295|issue:45701", "issue:43526|issue:43881", "issue:44671|issue:44743", - "issue:45081|issue:45812", "issue:41720|issue:45161", + "issue:45081|issue:45812", "issue:43531|issue:43572", "issue:43540|issue:43782", "issue:36010|issue:42617", @@ -3971,220 +3887,220 @@ "issue:44186|issue:44361", "issue:42915|issue:43322" ], - "prepared_review_unit_hash": "93591e487e18940438009a48340dfe8f71dff15331af637117d6e88be61690aa", + "prepared_review_unit_hash": "1d9946bb5e5a04a38c68fc0fa4f4f71823399f10b4fb9e95eb04324b7718722b", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36458, - "estimated_input_tokens": 9115, - "estimated_eval_tokens": 18486 + "serialized_chars": 36462, + "estimated_input_tokens": 9116, + "estimated_eval_tokens": 18488 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a noisy mix of unrelated Transformers issues. The only loose commonality is that several items involve tokenizer/processor regressions, but none of the soft pairs look like true duplicates or the same concrete fix.", - "confidence": 0.56, - "canonical_issue_reason": "issue:45701 is the broadest representative item: it is an open, high-interest tokenizer regression with multiple inbound references and best matches the small tokenizer-related subset of this cluster.", + "summary": "This cluster is mostly heterogeneous: the soft pairs share broad surface themes like tokenizers, loading, or offloading, but they do not look like the same underlying bug. I rejected all soft edges and picked the most connected issue only as a weak triage anchor, not as a true duplicate canonical.", + "confidence": 0.47, + "canonical_issue_reason": "issue:42617 is the most connected node in the soft-link graph, so it is the best weak anchor for triage, but its neighbors point to different failure modes and it is not a true duplicate center.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the best global issue candidate because it is broad, active, and could plausibly stand in for the tokenizer-regression theme better than the more model-specific reports.", + "best_issue_reason": "issue:42617 is the most central issue among this noisy set; it is the best representative for manual review, though not a safe duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Different bugs: cache-path collision for trust_remote_code vs permission propagation during save_pretrained. Same general custom-model area, but not the same code path." + "reason": "Cache-path collisions for trust_remote_code and read-only permission propagation are different bugs." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "Unrelated failures: a script launch problem vs a Qwen3Moe FP8 loading/config issue." + "reason": "3d_parallel.py execution failure and Qwen3Moe FP8 loading failure are unrelated code paths." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "Perplexity/evaluation anomaly is not the same as a tokenizer-version regression, even if tokenization could be a contributing factor." + "reason": "High perplexity on WikiText-2 and a tokenization regression are not the same failure." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "Tokenizer crash on NER/padding is unrelated to bfloat16 dtype mismatches." + "reason": "LayoutLMv2 tokenizer crashes and bf16 dtype mismatches have different root causes." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both involve serving multimodal models, but they are different crashes with different model/processor failures and no clear shared code-path bug." + "reason": "A continuous-batching multimodal crash with str.to and a missing _tokenizer on Gemma4Processor are distinct bugs." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "Different subsystems: Blip2 dtype loading vs Trainer resume/max_steps logic." + "reason": "Blip2 dtype handling and Trainer resume max_steps calculation are unrelated." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Both are memory-related, but one is int4 quantization/OOM and the other is an apex RMSNorm leak in T5; not the same fix." + "reason": "Int4 quantization memory growth and an apex RMSNorm memory leak are different issues." }, { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Both mention offloading/OOM, but the failures are about different mechanisms and platforms; too broad to treat as duplicates." + "reason": "Generic quantization offloading problems and Strix Halo mmap OOM are not the same bug." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Both are loading regressions, but one is strict config handling and the other is tokenizer loading for a specific model." + "reason": "Strict config loading and tokenizer loading failure are separate code paths." }, { "left": "issue:36331", "right": "issue:42617", "accept": false, - "reason": "Compute-loss API breakage is unrelated to the 3d_parallel.py execution failure." + "reason": "CustomTrainer.compute_loss signature breakage is unrelated to 3d_parallel.py startup failure." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Both involve model execution/loading problems, but torch.compile+SDPA and accelerate unpacking of packed weights are distinct bugs." + "reason": "torch.compile SDPA failure in Bamba and packed-weight unpacking in BitNet are different." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Trainer callback signature breakage is unrelated to the Qwen3Moe FP8 issue." + "reason": "Trainer compute_loss API breakage and Qwen3Moe FP8 loading failure do not match." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Video input shape handling and tokenization-version changes are different problems." + "reason": "Batched video shape handling and version-dependent tokenization changes are not the same bug." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Both are image-resolution issues, but they affect different processors/models and do not look like the same underlying bug." + "reason": "OmDet-Turbo input-size mismatch and Perceiver non-default-resolution failure are similar only at a high level, not the same code path." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Tokenizer crash vs processor output-size mismatch; different code paths." + "reason": "Tokenizer crashes in LayoutLMv2 and image-size mismatch in OmDet-Turbo are unrelated." }, { "left": "issue:44479", "right": "issue:44918", "accept": false, - "reason": "Video-input regression in multimodal models is unrelated to SFT trainer embedding unpacking." + "reason": "Qwen video-input regression and Qwen3.5 SFT embedding unpacking are different failures." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Different offload/device-mismatch failures in different models; too broad to merge." + "reason": "SDPA/compile failure and CPU-offload device mismatch are separate issues." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "A800 cudaErrorAssert under auto device mapping is unrelated to 3d_parallel.py launch failure." + "reason": "Qwen3 auto device mapping cudaErrorAssert and 3d_parallel.py execution failure are unrelated." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Both are config/API inconsistencies, but they concern different config classes and different behavior." + "reason": "Gemma3TextConfig API mismatch and GPTNeoX rotary_pct reload bug are different config problems." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Both involve loading/offload, but one is Qwen3Moe FP8 and the other is dequantized model loading with gpu+cpu offload; not the same bug." + "reason": "Qwen3Moe FP8 loading and dequantized model offload failures are different quantization/loading paths." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Processor.tokenizer regression is a processor API issue, not the same as a tokenization-output regression across versions." + "reason": "Processor.tokenizer regression and tokenization changes are not the same underlying bug." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to glm-4v loading failure." + "reason": "BeitImageProcessorFast label reduction bug and glm-4v-9b loading failure are unrelated." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "Masked-LM prediction bug in CamemBERT and Qwen3.5 recurrent-state reset are unrelated." + "reason": "CamemBERT masked-LM regression and Qwen3_5 recurrent-state reset are distinct model bugs." }, { - "left": "issue:45081", - "right": "issue:45812", + "left": "issue:41720", + "right": "issue:45161", "accept": false, - "reason": "Both are tokenizer bugs, but one is a Mistral regex patch crash and the other is Granite token ID corruption; different code paths and fixes." + "reason": "Auto device mapping failure and GPT-OSS tensor-parallel failure are different model-parallelism issues." }, { - "left": "issue:41720", - "right": "issue:45161", + "left": "issue:45081", + "right": "issue:45812", "accept": false, - "reason": "Different Qwen/GPT-OSS parallelism failures; not the same underlying bug." + "reason": "Mistral regex patch crash and Granite token-id regression are both tokenizer bugs but not the same fix." }, { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to StableLmConfig pad_token_idx." + "reason": "Qwen3-MoE sliding-window behavior and StableLm pad_token_idx omission are unrelated." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Video input processing error and weight_only loading error are different multimodal/model-loading failures." + "reason": "Video-processing ValueError in Qwen3OmniMoe and weight_only loading error in Qwen3VL are different multimodal failures." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "GenerationMixin import regression is unrelated to 3d_parallel.py execution failure." + "reason": "GenerationMixin import error and 3d_parallel.py execution failure have no common code path." }, { "left": "issue:43421", "right": "issue:45812", "accept": false, - "reason": "Runtime special-token post-processor updates and Granite wrong token IDs are both tokenizer-related, but not the same concrete bug." + "reason": "Runtime post-processor updates for special tokens and Granite token-id regression are different tokenizer behaviors." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Different tokenizers and different failure modes; too generic to merge." + "reason": "LayoutLMv2Tokenizer NER/padding crash and MLukeTokenizer task AttributeError are separate tokenizer bugs." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "Qwen3Moe FP8 failure and Llava Next segmentation fault are distinct model-loading bugs." + "reason": "Qwen3Moe FP8 failure and Llava Next segfault are unrelated model-loading problems." } ] }, @@ -4262,8 +4178,8 @@ "issue:44492|issue:44509", "issue:33357|issue:35707", "issue:43421|issue:45701", - "issue:41720|issue:43746", "issue:43606|issue:43720", + "issue:41720|issue:43746", "issue:30064|issue:34634", "issue:43550|issue:43653", "issue:43873|issue:45704", @@ -4287,217 +4203,217 @@ "issue:45704|issue:45923", "issue:44918|issue:44936" ], - "prepared_review_unit_hash": "f329b87f2c3c735c2e9260de9be880c7c4a66f4a2fefd9fc416e673d2c733325", + "prepared_review_unit_hash": "ab08afc26cf4eb06793ff123305a7fe25bd97bc0b7807091c4826bf10e667441", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36086, - "estimated_input_tokens": 9022, - "estimated_eval_tokens": 18300 + "serialized_chars": 36080, + "estimated_input_tokens": 9020, + "estimated_eval_tokens": 18296 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. The strongest link is the high-similarity pair issue:44514 / issue:44545, which appears to be the same batched `apply_chat_template` crash with `padding=False`. Most other soft edges look same-family or subsystem-adjacent, but not the same concrete bug and should remain separate.", - "confidence": 0.84, - "canonical_issue_reason": "issue:44514 is the best canonical representative because it is the earlier, more detailed report of the batched `Qwen2_5_VLProcessor.apply_chat_template` crash with `padding=False`; issue:44545 closely matches it and reads like a near-duplicate report of the same failure mode.", + "summary": "The cluster is mostly heterogeneous. Only one soft pair is a clear duplicate: the Qwen2_5_VLProcessor.apply_chat_template crash on batched input with padding=False.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 44545 is the best canonical issue because it is an exact duplicate of 44514, has the clearer later report, and directly names the same Qwen2_5_VLProcessor padding=False crash.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44514 is the best issue representative for this likely duplicate pair because it is the earliest and clearest description of the shared crash scenario, while the rest of the cluster does not support collapsing into one broader issue.", + "best_issue_reason": "Issue 44545 is the strongest issue candidate overall in this cluster because it is the only clear duplicate target and most precisely represents the shared bug in the accepted pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Different bugs: gather_for_metrics label truncation vs SAM3 text/bounding-box behavior." + "reason": "Different problems and code paths: metric gathering drops tuple labels vs SAM3 text/bbox behavior." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Different code paths and failures: missing `_tokenizer` on Gemma4Processor vs Qwen3.5 streaming inference crash." + "reason": "Different model/runtime failures: Gemma4Processor missing _tokenizer vs Qwen3.5 chat streaming API failure." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "Same documentation area, but these are distinct issues: a cache-strategy typo vs removed pipeline-task docs." + "reason": "One is a typo in cache strategy docs, the other is stale pipeline docs; related documentation cleanup but not the same bug." }, { "left": "issue:33357", "right": "issue:35707", "accept": false, - "reason": "Different failure modes: Mac CLIP bus error vs progressive generation with `inputs_embeds` and `past_key_values`." + "reason": "Unrelated failures: MacOS bus error with a CLIP model vs progressive generation with inputs_embeds/past_key_values." }, { "left": "issue:43421", "right": "issue:45701", "accept": false, - "reason": "Same tokenization theme, but one is runtime special-token post-processing and the other is a broader version-change report." + "reason": "Runtime special-token post-processor update bug is different from a broad tokenization regression report." }, { - "left": "issue:41720", - "right": "issue:43746", + "left": "issue:43606", + "right": "issue:43720", "accept": false, - "reason": "Different problems: Qwen3 auto device mapping cuda assert vs GraniteSpeech PEFT local checkpoint loading." + "reason": "Both involve loading/offload, but they are different concrete failures in different models: CPU offload device mismatch vs packed-weight unpacking." }, { - "left": "issue:43606", - "right": "issue:43720", + "left": "issue:41720", + "right": "issue:43746", "accept": false, - "reason": "Both involve loading/offload, but the concrete bugs differ: CPU-offload device mismatch vs packed-weight unpacking during accelerate loading." + "reason": "Different subsystems and failures: Qwen3 auto device mapping cuda assert vs GraniteSpeech PEFT local checkpoint loading." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Different processors and symptoms: void segmentation-map handling vs BarkProcessor `voice_preset`." + "reason": "Different processors and failures: void segmentation maps vs Bark voice_preset behavior." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "Different concrete issues: torch.compile/SDPA failure vs tokenizer mask-token registration." + "reason": "Different areas entirely: torch.compile/SDPA issue vs tokenizer special-token registration." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Different subsystems and root causes: quantization offloading vs T5/apex RMSNorm memory leak." + "reason": "Quantization/offloading behavior is unrelated to a T5 apex FusedRMSNorm memory leak." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same exact title and same concrete batched `apply_chat_template` crash with `padding=False`; this is the only strong duplicate pair here." + "reason": "Same exact Qwen2_5_VLProcessor.apply_chat_template batched-input crash with padding=False; clearly the same underlying bug." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Related to classification config, but the bugs are different: `num_labels` propagation in Qwen3.5 vs zero-loss behavior when `num_labels=1`." + "reason": "Both mention classification settings, but one is config propagation and the other is degenerate loss computation." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Different bugs: CPU-offload device mismatch vs tokenizer special-token decode output." + "reason": "Different failure modes and models; one is bark-small CPU offload mismatch, the other is a tokenizer special-token bug." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Same model family, but different concrete failures: sliding-window behavior in Qwen3-MoE vs `from_config` text-config unwrapping." + "reason": "Qwen3-MoE sliding_window behavior is not the same as unwrapping text_config for composite multimodal configs." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Different paths: trailing-slash local-path import warning vs undefined symbols in the video branch of multimodal token counting." + "reason": "A trailing-slash import warning is unrelated to undefined helpers in multimodal token counting." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "Different model failures: BarkProcessor `voice_preset` vs Qwen3 auto device mapping cudaErrorAssert." + "reason": "Different products and code paths: BarkProcessor voice_preset vs Qwen3 auto device mapping cudaErrorAssert." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token counting is unrelated to progressive generation with `inputs_embeds` and `past_key_values`." + "reason": "TrainerState token accounting is unrelated to progressive generation with embeddings and cache." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Both involve tokens, but one is runtime post-processor refresh and the other is save/load losing custom token IDs." + "reason": "Both involve special-token handling, but one is runtime post-processor refresh and the other is config save/load losing custom token IDs." }, { "left": "issue:44368", "right": "issue:45357", "accept": false, - "reason": "Same model family, different bugs: tie-word-embeddings warning vs incorrect visual encoder keys saved in `save_pretrained`." + "reason": "A warning about tie_word_embeddings is unrelated to incorrect visual encoder keys in save_pretrained." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Both Qwen-VL-related, but the failures differ: `weight_only=True` load error vs wrong video `vision_position_ids`." + "reason": "Different Qwen VL bugs: from_pretrained weight_only handling vs incorrect video vision_position_ids." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Different issues: Perceiver interpolation/resolution failure vs dtype mismatch in bfloat16 inference." + "reason": "Different subsystems: Perceiver image resolution interpolation vs dtype mismatches in SwitchTransformers/TimmWrapperModel inference." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Both are loading failures, but they concern different artifacts and root causes: model loading vs tokenizer loading." + "reason": "Both are model-loading/tokenizer issues, but they involve different models and different concrete tokenizer failures." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "Completely different topics: `patchtsmixer` post-init policy vs Qwen3.5 `num_labels` propagation." + "reason": "PatchTSMixer post_init policy and Qwen3.5 num_labels propagation are unrelated." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Different bugs: loading a specific model on transformers>=5.1 vs PIL backend processors requiring torchvision." + "reason": "Different loading failures: a specific model regression vs PIL backend image processors requiring torchvision." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Same broader Qwen3.5 family, but different concrete failures: composite config unwrapping vs CPU-offload rotary-embedding incompatibility." + "reason": "Different Qwen 3.x issues: config unwrapping vs CPU-offload incompatibility in rotary embedding." }, { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Unrelated failures: Qwen3-MoE sliding window behavior vs float8 storage lookup error." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to missing Float8 storage support." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "Different problems: `AddedToken` keyword conflict vs model loading failure." + "reason": "AddedToken argument duplication is a tokenizer construction bug, not the same as a model load failure for ScandiBERT." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "Unrelated: T5/apex memory leak vs Nemotron trust_remote_code gradient-checkpointing flag." + "reason": "A T5 apex RMSNorm memory leak is unrelated to a missing supports_gradient_checkpointing flag." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "Both trainer-related, but they describe different failures: TRL unpacking of Qwen3.5 input embeddings vs `evaluate()` after `train()`." + "reason": "Both are trainer-related, but one is unpacking Qwen3.5 input embeddings in TRL SFT and the other is evaluate() after train() failing." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The duplicate judgment for issue:44514/44545 is plausible, but the writeup overstates the evidence by calling it the \"only confirmed duplicate pair\" and by implying the rest are simply unrelated. The packet only supports a high-similarity/likely-duplicate reading for 44514/44545; it does not confirm duplicates elsewhere, and several rejected pairs are at least same-family rather than clearly unrelated. Please soften the canonicality language and avoid confirmation claims not present in the packet." + "accept": true, + "feedback": "The summary is grounded in the packet, and the one accepted pair (44514/44545) is supported by the identical issue title and matching crash description. The other soft-edge verdicts are conservative and avoid overstating similarity." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -4566,8 +4482,8 @@ ], "soft_pairs": [ "issue:39692|issue:44610", - "issue:45663|issue:45923", "issue:43421|issue:43653", + "issue:45663|issue:45923", "issue:45588|issue:45759", "issue:42886|issue:44843", "issue:43381|issue:43844", @@ -4576,11 +4492,11 @@ "issue:43525|issue:43531", "issue:45704|issue:45859", "issue:45081|issue:45356", - "issue:45356|issue:45907", "issue:39692|issue:43873", + "issue:45356|issue:45907", "issue:43746|issue:43873", - "issue:40444|issue:43746", "issue:43596|issue:44220", + "issue:40444|issue:43746", "issue:43746|issue:45381", "issue:43540|issue:44008", "issue:43720|issue:43819", @@ -4596,221 +4512,224 @@ "issue:43550|issue:43819", "issue:43322|issue:45685" ], - "prepared_review_unit_hash": "85860d8b06f66e42e79772c44f3c09c379d34b3fe713d07593f3d162ca9b1ab7", + "prepared_review_unit_hash": "40a6c2843afed04e8fc0fba671695515574bc815e6475e0a95cbaea95a79152e", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 37187, - "estimated_input_tokens": 9297, - "estimated_eval_tokens": 18850 + "serialized_chars": 37181, + "estimated_input_tokens": 9296, + "estimated_eval_tokens": 18848 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a set of mostly unrelated issue reports with only superficial lexical overlap. None of the candidate edges look like true duplicates or the same underlying fix, so all soft edges should be rejected. No single issue or PR stands out as a canonical duplicate target.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "The set is largely heterogeneous; most soft pairs are separate bugs touching different models or code paths. The only convincing duplicate pair is the HF_HUB_OFFLINE / tokenizer cache failure around `_patch_mistral_regex` (`issue:42886` and `issue:44843`).", + "confidence": 0.9, + "canonical_issue_reason": "For the only clear duplicate subcluster, `issue:44843` is the better canonical because it identifies the concrete root cause (`_patch_mistral_regex` unconditionally calling `model_info()` in offline mode) rather than only the broader offline-cache symptom.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "`issue:44843` is the clearest representative of the one true duplicate pair in this set; the rest of the issues are too unrelated to share a single canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both are vision/doc example mismatches, but one is SigLIP2 example errors/quantization and the other is OmDet-Turbo 640px vs 224px input size; different bugs and paths." + "reason": "Both are vision-model example failures, but one is a SigLIP2 doc/example mismatch and the other is an OmDet-Turbo processor/model size mismatch; not the same bug." }, { - "left": "issue:45663", - "right": "issue:45923", + "left": "issue:43421", + "right": "issue:43653", "accept": false, - "reason": "Gemma-4 FSDP2 KeyError in attention state rebuilding vs Nemotron trust_remote_code gradient-checkpointing flag missing; unrelated failures." + "reason": "Both involve tokenizers and special tokens, but one is about runtime post-processor updates and the other about BigBird mask-token registration/decoding; different defects." }, { - "left": "issue:43421", - "right": "issue:43653", + "left": "issue:45663", + "right": "issue:45923", "accept": false, - "reason": "Both involve tokenizer special tokens, but one is runtime post-processor updates and the other is BigBirdTokenizer mask-token registration/decode behavior; not the same bug." + "reason": "Different problems: a Gemma4 FSDP2 per-layer `shared_kv_states` rebuild bug versus a missing `supports_gradient_checkpointing` flag on a trust_remote_code variant." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Different code paths: flash_attention s_aux=None crash vs AutoModelForCausalLM.from_config not unwrapping text_config." + "reason": "One is a flash-attention crash on `s_aux=None`; the other is a composite Qwen config unwrap issue in `from_config`. No shared concrete code-path bug." }, { "left": "issue:42886", "right": "issue:44843", - "accept": false, - "reason": "Both touch offline behavior, but one is cache loading under HF_HUB_OFFLINE and the other is an unconditional model_info() call in _patch_mistral_regex; distinct issues." + "accept": true, + "reason": "Same underlying offline-tokenizer failure: both report HF_HUB_OFFLINE/cache-loading breakage, and `issue:44843` identifies the concrete `_patch_mistral_regex`/`model_info()` cause." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Gradient checkpointing eval-mode validation vs ZeRO-3 gradient growth during training; different mechanisms and symptoms." + "reason": "Both mention training/runtime behavior, but one is gradient checkpointing in eval mode and the other is gradient explosion with ZeRO-3; not duplicates." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Runtime special-token/post-processor update bug vs LayoutLMv2Tokenizer crashes on NER/batched padding; not the same underlying defect." + "reason": "Tokenizer-related, but different failure modes: post-processor not updating special tokens versus LayoutLMv2 NER/batched padding crashes." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different model families and outputs: SigLIP2 AutoModel/pipeline nonsense vs CamemBERT masked LM prediction regression." + "reason": "Different model regressions with different symptoms and code paths; no evidence they are the same bug." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Llama4Config pad_token_id AttributeError and Qwen3-MoE sliding_window handling are unrelated configuration bugs." + "reason": "One is a missing `pad_token_id` attribute on Llama4Config; the other is a Qwen3-MoE `sliding_window` issue. Distinct." }, { "left": "issue:45704", "right": "issue:45859", "accept": false, - "reason": "T5 apex FusedRMSNorm memory leak vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility; no shared fix path." + "reason": "T5/apex RMSNorm memory leak and Qwen3_5Moe CPU-offload incompatibility are unrelated." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer-regex related, but one is an AttributeError in _patch_mistral_regex and the other is a Kimi-K2.5 codec/regression warning issue; not duplicates." + "reason": "Both touch Mistral-related tokenizer logic, but one is a hard crash in `_patch_mistral_regex` and the other is a codec-handling/regression report with a warning issue; not the same concrete bug." }, { - "left": "issue:45356", - "right": "issue:45907", + "left": "issue:39692", + "right": "issue:43873", "accept": false, - "reason": "Kimi tokenizer regression vs list_repo_files missing revision forwarding in CI; unrelated components." + "reason": "A doc/example mismatch is not the same as a quantization/offloading behavior bug." }, { - "left": "issue:39692", - "right": "issue:43873", + "left": "issue:45356", + "right": "issue:45907", "accept": false, - "reason": "SigLIP2 doc example errors/quantization failure vs quantization offloading behavior; too broad and different concrete bugs." + "reason": "Kimi tokenizer regression and `list_repo_files` not forwarding `revision` are unrelated." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "GraniteSpeech PEFT local-checkpoint loading vs quantization offloading; different subsystems and failure modes." + "reason": "PEFT/local checkpoint loading for GraniteSpeech and quantization offloading behavior are different problems." }, { - "left": "issue:40444", - "right": "issue:43746", + "left": "issue:43596", + "right": "issue:44220", "accept": false, - "reason": "Qwen2.5-VL iterable-dataset multi-image finetuning failure and GraniteSpeech PEFT checkpoint loading are unrelated." + "reason": "Deepspeed ZeRO-3 initialization with BertModel and `_torch_extract_fbank_features()` are unrelated code paths." }, { - "left": "issue:43596", - "right": "issue:44220", + "left": "issue:40444", + "right": "issue:43746", "accept": false, - "reason": "DeepSpeed zero3/BertModel index error vs _torch_extract_fbank_features issue; different code paths entirely." + "reason": "Qwen2.5-VL iterable dataset/multi-image finetuning and GraniteSpeech PEFT checkpoint loading are different bugs." }, { "left": "issue:43746", "right": "issue:45381", "accept": false, - "reason": "GraniteSpeech PEFT loading bug vs Qwen2.5-VL video vision_position_ids bug; separate multimodal problems." + "reason": "PEFT/local checkpoint loading and Qwen2.5-VL video `vision_position_ids` are unrelated." }, { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Qwen3OmniMoe video processing ValueError vs Gemma3n audio_mel_mask variable collision; unrelated multimodal bugs." + "reason": "Qwen3OmniMoe video-processing ValueError and Gemma 3n `audio_mel_mask` name-collision AttributeError are different defects." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "BitNet packed-weight loading bug vs DAC.from_latents vs forward mismatch; different models and code paths." + "reason": "BitNet accelerate-loading unpacking issue and DAC `from_latents`/STE mismatch are unrelated." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Audio feature extraction issue vs add-new-model-like failing in TOKENIZER_MAPPING_NAMES; unrelated." + "reason": "An audio feature-extraction issue and an `add-new-model-like` tokenizer-mapping failure are not the same bug." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Input_ids/inputs_embeds validation logic bug vs zero-loss classification when num_labels=1; different behavior and fixes." + "reason": "Wrong `input_ids`/`inputs_embeds` validation messaging is separate from the single-label-classification zero-loss bug." }, { "left": "issue:43618", "right": "issue:43697", "accept": false, - "reason": "CLIPOutput attentions assignment regression vs RTDetrV2 different outputs in v5; not the same bug." + "reason": "CLIP attentions not assigned and RTDetrV2 output differences in v5 are different model regressions." }, { "left": "issue:30064", "right": "issue:36010", "accept": false, - "reason": "Image processor void-segmentation-map failure and GenerationMixin import error are completely unrelated." + "reason": "Image processor void-segmentation-map failure and the `GenerationMixin` import error are unrelated." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "BigBird special-token decode bug vs BitNet packed-weights accelerate loading bug; no common underlying change." + "reason": "BigBird special-token registration and BitNet packed-weight loading are entirely different issues." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "SwitchTransformers sparse-layer creation bug vs Smollm3 RoPE-layer drop bug; different model architecture issues." + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are unrelated model-config bugs." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Mistral-3 image preprocessor/model dtype mismatch vs Gemma3n audio tensor name collision; unrelated." + "reason": "Mistral-3 image preprocessor dtype mismatch and Gemma 3n audio-mask name collision are different problems." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Processor/tokenizer regression and Kosmos2.5 OCR long-input index error are different multimodal bugs." + "reason": "A processor/tokenizer regression and a Kosmos2.5 long-OCR index error do not share the same underlying cause." }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Bamba torch.compile+SDPA failure vs SwitchTransformers/TimmWrapperModel bfloat16 dtype mismatch; distinct issues." + "reason": "Bamba torch.compile/SDPA failure and dtype mismatches in SwitchTransformers/TimmWrapperModel are distinct." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Bamba compile/SDPA problem and DAC.from_latents mismatch are unrelated." + "reason": "Bamba SDPA/torch.compile issue and DAC latent conversion mismatch are unrelated." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Llava Next loading segfault vs MPS histogram_mps not implemented for Int; different failure classes and models." + "reason": "Llava Next segfault and MPS histogram int support error are not the same bug." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The analyst stays conservative and grounds the summary in the packet. The single accepted pair, issue:42886 and issue:44843, is plausibly the same offline-tokenizer / `_patch_mistral_regex` failure, and the other soft pairs are reasonably rejected as distinct bugs rather than over-merged." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -4822,8 +4741,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -4904,196 +4823,196 @@ "issue:43606|issue:45072", "issue:44671|issue:45701" ], - "prepared_review_unit_hash": "1fe739a34374f5cc03c7937fe57279b98bc84dd39808ca7ef28b18c87d022a3a", + "prepared_review_unit_hash": "302a991f7d2001395c60bed868515b7b351c9731cb7178af2811b38e368edab4", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35262, - "estimated_input_tokens": 8816, - "estimated_eval_tokens": 17888 + "serialized_chars": 35265, + "estimated_input_tokens": 8817, + "estimated_eval_tokens": 17890 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is dominated by unrelated regression reports around tokenizers, configs, vision/video models, and v5 behavior changes; there are no clear duplicate pairs with the evidence provided. The best single representative is the broad remote-code regression report.", - "confidence": 0.44, - "canonical_issue_reason": "issue:45020 is the broadest and most representative umbrella report in the set, covering recent-version breakage for remote_code models without being overly tied to one specific model.", + "summary": "The cluster is mostly a set of unrelated bug reports that only share broad themes like tokenizer, multimodal, and v5 regression language. The soft pairs are mostly superficial similarities and do not look like the same underlying bug or change.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45020 is the broadest open umbrella here, covering recent version regressions around remote_code/config loading; the other issues are narrower, model-specific failures.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 best captures the cluster\u2019s overall theme of Transformers version regressions affecting model loading/runtime behavior across multiple remote-code scenarios.", + "best_issue_reason": "issue:45020 is the best anchor for the cluster because it is the most general and central regression report, even though it does not duplicate the other issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "Different failures: AddedToken keyword collision vs tokenizer loading failure for a specific model." + "reason": "Different failure modes: AddedToken constructor error vs tokenizer/model loading failure for a specific repo." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Both are image/tokenization bugs, but they affect different models and code paths." + "reason": "Different models and code paths: LayoutLMv2 tokenizer padding/NER crash vs Perceiver image classification resolution/interpolation bug." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Same helper function area, but one is an offline metadata call bug and the other is an AttributeError in regex patching." + "reason": "Both touch _patch_mistral_regex, but one is an offline model_info call and the other is an AttributeError on tokenizer internals; not the same bug." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Unrelated problems: Qwen3-MoE sliding window behavior vs config fields missing under trust_remote_code." + "reason": "Unrelated areas: Qwen3-MoE sliding_window behavior vs AutoConfig trust_remote_code field loss." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "BigBird tokenizer special-token registration and DAC latent reconstruction are unrelated." + "reason": "Tokenizer special-token registration bug vs DAC latents/STE forward mismatch; no common code-path." }, { "left": "issue:43577", "right": "issue:43818", "accept": false, - "reason": "Different model families and different defects: dtype loading vs missing temporal attention/shared weights." + "reason": "Different multimodal models and defects: dtype propagation in Blip2 vs missing temporal attention / shared weights in Video-LLaVA." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "Both involve config loading, but the reported failures are different and not the same bug." + "reason": "Both involve config loading, but one is missing fields from trust_remote_code and the other is strict config rejection for granite_speech." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Different video-input regressions in different Qwen Omni/VL variants." + "reason": "Different Qwen video bugs: Qwen3OmniMoe video processing ValueError vs Qwen2.5-VL vision_position_ids regression." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Runtime special-token/post-processor update bug vs Whisper batch_decode skip_special_tokens handling." + "reason": "Runtime special-token/post-processor mutation issue vs Whisper batch_decode skip_special_tokens handling; different components." }, { "left": "issue:43782", "right": "issue:44977", "accept": false, - "reason": "Qwen3VL weight_only loading failure is unrelated to flash-attention generation issues." + "reason": "Loading error with weight_only=True vs generation failure with flash-attention; not the same concrete bug." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Different regressions: Pegasus output changes across versions vs CamemBERT masked LM predictions." + "reason": "Both are v5 regressions, but one is Pegasus output drift and the other is CamemBERT masked-LM prediction corruption; different models and causes." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation and Kosmos2.5 long-OCR index errors are unrelated." + "reason": "Whisper empty transcription generation vs Kosmos2.5 long OCR index error; different models and failure signatures." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "Pipeline error-message wording and SigLIP2 model/pipeline mismatch are different issues." + "reason": "Docs/error-message wording around removed pipeline tasks vs a SigLIP2 model/pipeline output problem; not the same bug." }, { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Both involve remote-code breakage, but one is specifically the removal of is_torch_fx_available while the other is a broader umbrella report." + "reason": "Related only at a broad level: 44561 is a specific remote_code break from removing is_torch_fx_available, while 45020 is a generic umbrella report." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "SigLIP2 docs/model mismatch and Kosmos2.5 OCR indexing are unrelated." + "reason": "SigLIP2 docs/example errors are unrelated to Kosmos2.5 OCR index failures." }, { "left": "issue:44485", "right": "issue:44509", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to v5 pipeline doc references." + "reason": "GLM-5 RoPE implementation discussion vs outdated docs about removed pipeline tasks; different issues." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "AutoImageProcessor-from-URL loading and template compilation errors are different failure modes." + "reason": "AutoImageProcessor URL loading vs Jinja/template compilation TypeError; unrelated subsystems." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "ImageProcessor void segmentation maps and Trainer compute_loss signature mismatch are unrelated." + "reason": "Image processor segmentation-map failure vs trainer.compute_loss signature mismatch; no overlap in code path." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Qwen3 auto device mapping CUDA assert vs tokenizer loading failure for a different model." + "reason": "Qwen3 auto device-map CUDA assert vs tokenizer loading regression for EMBEDDIA/est-roberta; different problems." }, { "left": "issue:43644", "right": "issue:43749", "accept": false, - "reason": "Non-persistent buffer junk filling and FSDP CPU RAM efficient loading are not the same underlying bug." + "reason": "Non-persistent buffer initialization bug vs FSDP CPU RAM efficient loading breakage; related only by version." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes and AutoTokenizer failing on FastSpeech2ConformerTokenizer are distinct tokenizer bugs." + "reason": "Tokenizer crash in LayoutLMv2 vs AutoTokenizer loading failure for FastSpeech2Conformer; different artifacts and code paths." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Both relate to v5 pipeline removals, but one is an error-message issue and the other is stale docs; not the same concrete bug." + "reason": "Both reference removed v5 pipeline tasks, but one is a runtime message issue and the other is stale docs; not the same bug/change." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Qwen video-input regression and Kosmos2.5 OCR index error affect different models and code paths." + "reason": "Qwen video-input regression vs Kosmos2.5 OCR long-input index error; unrelated model families." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "Blip2 dtype loading and Apple Silicon caching_allocator_warmup TypeError are unrelated." + "reason": "Blip2 dtype propagation issue vs Apple Silicon caching_allocator_warmup TypeError; distinct failures." }, { "left": "issue:44805", "right": "issue:44936", "accept": false, - "reason": "Mask/tensor shape IndexError and trainer evaluate-after-train behavior are different problems." + "reason": "Tensor mask shape mismatch vs trainer.evaluate after train lifecycle bug; different concrete problems." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "CPU offload device mismatch for bark-small and bfloat16 inference dtype mismatch are unrelated." + "reason": "CPU offload device mismatch in bark-small vs bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "CamemBERT masked LM regression and generic tokenization changes across versions are not the same bug." + "reason": "CamemBERT v5 prediction regression vs general tokenization changes across versions; not the same bug." } ] }, @@ -5112,8 +5031,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -5192,206 +5111,206 @@ "issue:38175|issue:44610", "issue:44589|issue:45042", "issue:43854|issue:44977", - "issue:39692|issue:44912", "issue:43819|issue:43927", + "issue:39692|issue:44912", "issue:43257|issue:43329" ], - "prepared_review_unit_hash": "1eb830627c82a96a987667d73ff6a3e9d8c6702317bd1aae0dffa9a48a66c6ed", + "prepared_review_unit_hash": "3cfa70e286fe3ffcd2c7c02b2e87b8a7628b57cd9fe1d17b595c5a62bf7a38fb", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36473, - "estimated_input_tokens": 9119, - "estimated_eval_tokens": 18494 + "serialized_chars": 36472, + "estimated_input_tokens": 9118, + "estimated_eval_tokens": 18492 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a collection of mostly unrelated Transformers bug reports; the soft pairs share only broad subsystem similarity, not the same underlying defect. No soft edge should be merged, and there is no clear duplicate PR candidate here.", - "confidence": 0.94, - "canonical_issue_reason": "issue:45020 is the broadest open regression report in the set and the most plausible umbrella issue; the rest are narrower, model- or component-specific failures that do not collapse into one concrete bug.", + "summary": "Mostly unrelated Transformers bug reports spanning model loading, tokenizers, processors, quantization, and training. None of the soft pairs look like the same underlying defect, so all candidate merges are rejected. The broadest representative issue is the remote_code regression report.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45020 is the broadest umbrella report about recent Transformers versions breaking remote_code models, so it best represents this mixed regression cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the best representative issue because it is general, recent, and framed as a cross-cutting regression affecting multiple models via remote_code loading.", + "best_issue_reason": "issue:45020 is the most general and centrally scoped issue in the set; the others are narrower, model-specific failures rather than a shared duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "Different failures: BarkProcessor voice_preset behavior vs a GenerationMixin import path break. Same library, unrelated bugs." + "reason": "Different subsystems and failure modes: BarkProcessor voice_preset vs GenerationMixin import regression." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "OmDet processor image-size mismatch is unrelated to SwitchTransformers/TimmWrapperModel bfloat16 dtype mismatches." + "reason": "Different bug classes: processor image size mismatch vs dtype mismatch in inference." }, { "left": "issue:43782", "right": "issue:45325", "accept": false, - "reason": "Qwen3VL weight_only load failure and Qwen2.5-VL rope index scaling bug affect different model paths and different code paths." + "reason": "Both are model-loading related, but one is weight_only loading and the other is a rope-index temporal-position bug." }, { "left": "issue:43720", "right": "issue:45072", "accept": false, - "reason": "BitNet packed-weight loading during accelerate is not the same bug as bfloat16 dtype mismatch in other models." + "reason": "BitNet packed-weight loading failure is unrelated to bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Backend dependency packaging issue vs Mistral tokenizer regex patch crash; unrelated root causes." + "reason": "Packaging/backend dependency issue vs tokenizer regex patch crash; not the same defect." }, { "left": "issue:45081", "right": "issue:45406", "accept": false, - "reason": "Both are tokenizer/processor loading issues, but one is a regex patch crash and the other is a missing processor attribute in serve; different bugs." + "reason": "Both involve tokenizer/processor code, but different objects and crashes in different call paths." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM efficient loading and init_empty_weights _is_hf_initialized argument errors are separate loading mechanisms and distinct regressions." + "reason": "Both concern loading flows, but one is FSDP CPU RAM-efficient loading and the other is init_empty_weights argument handling." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "Low-quality placeholder issue vs an isolated reproduction of another issue; no shared underlying bug is evident." + "reason": "Generic placeholder issue vs isolated reproduction of a specific external bug; not duplicates." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Both mention tied-weights warnings, but one is a Qwen3-Next weight-tying bug and the other is a user-level warning during LoRA fine-tuning; not the same concrete defect." + "reason": "Different tied-weights warnings affecting different model families and configurations." }, { "left": "issue:43582", "right": "issue:44857", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError and CUDA float16 AMP crash in LwDetrImageLoss are unrelated runtime failures." + "reason": "Apple Silicon allocator TypeError is unrelated to CUDA AMP crash in a detection loss." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting bug is unrelated to DAC.from_latents not matching forward pass." + "reason": "Audio sampling-rate default bug is unrelated to DAC latent/forward-pass mismatch." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BEiT image label reduction bug and RTDetrV2 output divergence are different models and different code paths." + "reason": "BEiT reduce_labels bug and RTDetrV2 output drift are separate model-specific defects." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "BEiT label reduction and an MPS attention correctness issue do not describe the same underlying defect." + "reason": "No shared code path: BEiT label reduction vs MPS attention correctness on differing head dims." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve exception handling, but one hides tokenizer decode errors without protobuf and the other is a bare except in Fuyu tensor conversion; different code paths." + "reason": "Both involve error handling, but one is protobuf decode error masking and the other is a bare except in tensor conversion." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "BarkProcessor voice_preset failure is unrelated to CustomTrainer.compute_loss signature mismatch." + "reason": "Unrelated: BarkProcessor preset behavior vs CustomTrainer compute_loss signature change." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Template compilation TypeError and CLIPTokenizer model_max_length sentinel value are unrelated bugs." + "reason": "Template compilation error and CLIPTokenizer max length bug are unrelated." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Undefined variables in multimodal token counting are unrelated to stale auto_mappings references breaking CI." + "reason": "Multimodal token counting bug is unrelated to stale auto_mappings config references." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "KeyError loading a specific MobileLLM checkpoint is a model-specific loading bug, not the broader remote_code regression in 45020." + "reason": "Specific MobileLLM key error does not match the broader remote_code regression." }, { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "BigBirdTokenizer special-token decode bug and MXFP4 quantization fallback are different tokenizer/quantization issues." + "reason": "Tokenizer special-token registration bug is unrelated to MXFP4 quantization fallback." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Zero3 training IndexError on a non-pretrained Bert model is not the same as evaluate() failing after train()." + "reason": "Zero3/non-pretrained Bert index error is not the same as trainer.evaluate() failing after train()." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "GraniteSpeech PEFT local checkpoint loading and Whisper batch_decode skip_special_tokens handling are different processor/model behaviors." + "reason": "Local PEFT checkpoint loading issue is unrelated to Whisper batch_decode skip_special_tokens handling." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "gguf_file loader breakage and tokenizer loading failure for a specific model are both loading issues, but not the same code-path or defect." + "reason": "gguf_file support failure is different from tokenizer loading regression for est-roberta." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Zero probabilities in SigLIP2 and OmDet processor/model input-size mismatch are unrelated vision-model bugs." + "reason": "Unexpected zero probabilities in SigLIP2 are not the same as OmDet processor input-size mismatch." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Missing Float8 storage type and PIL backend requiring torchvision are separate dependency/runtime issues." + "reason": "Storage lookup failure for float8 is unrelated to torchvision dependency requirements in PIL image processors." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "GLM-4.7-Flash unit-test loading problem and Qwen3.5 flash-attention generation issue are distinct model-specific regressions." + "reason": "Model loading/test issue for GLM-4.7-Flash is not the same as flash-attention generation failure in Qwen3.5." }, { - "left": "issue:39692", - "right": "issue:44912", + "left": "issue:43819", + "right": "issue:43927", "accept": false, - "reason": "SigLIP2 documentation example errors and MXFP4 quantization fallback are not the same underlying bug." + "reason": "DAC latent mismatch and DiaConfig custom token ID loss are separate generation/model-config bugs." }, { - "left": "issue:43819", - "right": "issue:43927", + "left": "issue:39692", + "right": "issue:44912", "accept": false, - "reason": "DAC forward/from_latents mismatch and DiaConfig losing custom token IDs after save/load are unrelated state/serialization bugs." + "reason": "SigLIP2 doc example errors do not match MXFP4 quantization fallback behavior." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Qwen3 MOE weight conversion during accelerate+deepspeed is unrelated to undefined video token count helpers in multimodal processors." + "reason": "Qwen3 MOE weight conversion/loading issue is unrelated to multimodal token counting undefined variables." } ] }, @@ -5410,8 +5329,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -5477,8 +5396,8 @@ "issue:36010|issue:36331", "issue:44568|issue:45356", "issue:43749|issue:44756", - "issue:44568|issue:45488", "issue:43643|issue:45759", + "issue:44568|issue:45488", "issue:34567|issue:40990", "issue:43329|issue:43450", "issue:44849|issue:45701", @@ -5487,216 +5406,216 @@ "issue:43697|issue:45357", "issue:43257|issue:43756", "issue:44479|issue:45381", - "issue:44987|issue:45538", "issue:43452|issue:43582", + "issue:44987|issue:45538", "issue:41720|issue:43540", "issue:44291|issue:45042", - "issue:44987|issue:45907", "issue:42915|issue:43742", + "issue:44987|issue:45907", "issue:33357|issue:34567" ], - "prepared_review_unit_hash": "580e017ab238ce5adc1644b1da3dc4ba09a89524416c0cc1d64bf22c7739651c", + "prepared_review_unit_hash": "3282751fd1ac6e24f64b39d934f4103215310d01de41d429960799e10ff376e0", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36603, - "estimated_input_tokens": 9151, - "estimated_eval_tokens": 18558 + "serialized_chars": 36597, + "estimated_input_tokens": 9150, + "estimated_eval_tokens": 18556 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is highly heterogeneous: tokenizer regressions, multimodal/video bugs, model-loading issues, trainer API breakages, and offload/device problems. I did not find any soft pair that looks like the same underlying bug/change.", - "confidence": 0.81, - "canonical_issue_reason": "Issue 44479 is the most concrete broad regression report in the set, covering multiple Qwen video-input paths and making it the best umbrella-style anchor among otherwise unrelated issues.", + "summary": "This cluster is heterogeneous: tokenization regressions, model-loading bugs, multimodal/video processor issues, and training API breakages. None of the soft pairs look like the same underlying bug/change, so I would not deduplicate them.", + "confidence": 0.94, + "canonical_issue_reason": "issue:45701 is the broadest open issue and best fits an umbrella for the tokenizer/version-regression subset, with multiple inbound references and a title that captures the general symptom pattern.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44479 is the strongest representative issue because it describes a specific regression with multiple affected model families and is more triage-useful than the more generic or highly model-specific reports.", + "best_issue_reason": "issue:45701 is the most central and reusable issue in this set; it\u2019s broad enough to represent the shared tokenizer-regression theme better than the highly specific model- or API-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "Different failures: a missing import for GenerationMixin vs. custom model initialization breaking in notebooks." + "reason": "Different breakages: a missing GenerationMixin import vs Jupyter-defined custom model initialization in v5." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "Unrelated subsystems: accelerate-loaded packed weights vs. image processor output size mismatch." + "reason": "Both are loading/vision-adjacent failures, but one is about unpacking packed BitNet weights during accelerate loading and the other about processor image size mismatch." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Both are model/quantization related, but they describe different concrete bugs on different models and code paths." + "reason": "Different concrete problems: torch.compile with SDPA on Bamba vs MXFP4 quantization fallback on gpt-oss-20b." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "Different model-specific loading issues: weight tying bug vs. dtype not propagating to BLIP2/Qformer." + "reason": "AyaVision weight tying bug is unrelated to Blip2 dtype propagation on model load." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "Loading/dispatch bug vs. CUDA OOM; not the same underlying problem." + "reason": "Qwen3 MOE weights not converted during accelerate+deepspeed is distinct from a plain CUDA OOM on gpt-oss-20b." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio chat template sampling-rate default vs. Whisper batch_decode special-token handling." + "reason": "Audio apply_chat_template sampling-rate default bug is unrelated to Whisper batch_decode skipping special tokens." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "Both involve video/multimodal paths, but one is an undefined-symbol bug in token counting while the other is a broader video-input regression." + "reason": "Undefined video token counting helpers vs a broader video-input regression in several VLMs are related by modality only, not the same bug." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "A local-path import warning is not the same as Qwen3 MoE weight conversion under accelerate+deepspeed." + "reason": "Local-path import warning with trailing slash is unrelated to Qwen3 MOE weight conversion under accelerate/deepspeed." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and PIL processor torchvision dependency are unrelated." + "reason": "Sliding window behavior in Qwen3-MoE is not the same as PIL-backed image processors wrongly requiring torchvision." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "Import path breakage and a Trainer compute_loss signature change are different API regressions." + "reason": "ImportError for GenerationMixin and unexpected num_items_in_batch in CustomTrainer are separate API regressions." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both tokenizer regressions, but they affect different tokenizers and distinct behaviors." + "reason": "Both tokenizer-related, but one is add_special_tokens/BOS-EOS behavior while the other is codec handling and warning text in Kimi-K2.5." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different memory/loading issues." + "reason": "FSDP CPU RAM efficient loading and disabling mmap on Strix Halo are different memory-loading issues with different triggers." }, { - "left": "issue:44568", - "right": "issue:45488", + "left": "issue:43643", + "right": "issue:45759", "accept": false, - "reason": "Both are tokenizer-related, but one is missing BOS/EOS insertion and the other is LlamaTokenizer overriding tokenizer JSON behavior." + "reason": "Missing fields from trust_remote_code AutoConfig is not the same as AutoModelForCausalLM.from_config failing to unwrap text_config in composite Qwen configs." }, { - "left": "issue:43643", - "right": "issue:45759", + "left": "issue:44568", + "right": "issue:45488", "accept": false, - "reason": "Config-loading with trust_remote_code missing fields is different from unwrapping text_config for composite multimodal configs." + "reason": "Different tokenizer regressions: add_special_tokens handling vs LlamaTokenizer overriding ByteLevel pre-tokenizer." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "Trainer token-count tracking and GPT-OSS perplexity are unrelated." + "reason": "TrainerState token counting not updating is unrelated to high perplexity on a specific model/dataset." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Undefined helpers in multimodal token counting is not the same as batched video processor shape output." + "reason": "Undefined helper in multimodal token counting vs batched video processor shape mismatch are separate video-path bugs." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "A Qwen3.5 hidden-states bug and a generic tokenization change report are not the same bug." + "reason": "Qwen3.5 output_hidden_states bug is model-specific and not the same as the broad tokenization-version regression." }, { "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "Different Qwen3 loading problems: deepspeed/accelerate weight conversion vs. shape mismatch for Qwen3-VL." + "reason": "Qwen3 MOE accelerate/deepspeed conversion failure is different from Qwen3-VL weight shape mismatch on model load." }, { "left": "issue:43606", "right": "issue:45859", "accept": false, - "reason": "Both mention CPU offload, but they affect different models and likely different offload code paths." + "reason": "CPU offload device mismatch in bark-small is unrelated to Qwen3_5MoeTextRotaryEmbedding CPU offload compatibility." }, { "left": "issue:43697", "right": "issue:45357", "accept": false, - "reason": "RTDetrV2 output drift and Qwen3.5 save_pretrained key corruption are unrelated." + "reason": "RTDetrV2 differing outputs in v5 is not the same as Qwen3.5 save_pretrained writing incorrect visual encoder keys." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Qwen3 MoE weight conversion and Smollm3 RoPE-layer dropping are different model bugs." + "reason": "Qwen3 MOE conversion/loading failure is unrelated to Smollm3 RoPE layer dropping." }, { "left": "issue:44479", "right": "issue:45381", "accept": false, - "reason": "Related video-input area, but the titles point to different symptoms and scopes; not enough to treat as the same underlying bug." + "reason": "General video regression across several models is too broad to merge with the specific qwen2.5-vl vision_position_ids issue." }, { - "left": "issue:44987", - "right": "issue:45538", + "left": "issue:43452", + "right": "issue:43582", "accept": false, - "reason": "Generic loading failure and CLIPTokenizer max-length behavior are unrelated." + "reason": "gguf_file breaks AutoTokenizer/AutoModel loading, while caching_allocator_warmup fails on Apple Silicon with TypeError; different code paths." }, { - "left": "issue:43452", - "right": "issue:43582", + "left": "issue:44987", + "right": "issue:45538", "accept": false, - "reason": "GGUF loader breakage and Apple Silicon warmup TypeError are different failures." + "reason": "Physical-intelligence/fast loading failure is unrelated to CLIPTokenizer's pathological model_max_length." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "CUDA device-mapping assert on Qwen3 is not the same as Qwen3OmniMoe video ValueError." + "reason": "Qwen3 auto device mapping cudaErrorAssert and Qwen3OmniMoe video input ValueError are different runtime failures." }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "init_empty_weights argument handling and PIL backend torchvision dependency are unrelated regressions." + "reason": "init_empty_weights _is_hf_initialized argument issue is unrelated to PIL backend processors requiring torchvision." }, { - "left": "issue:44987", - "right": "issue:45907", + "left": "issue:42915", + "right": "issue:43742", "accept": false, - "reason": "Loading physical-intelligence/fast and forwarding revision in list_repo_files are different repository API problems." + "reason": "Qwen3Moe FineGrainedFP8Config failure and MobileLLM-125M key error are distinct loading/config bugs." }, { - "left": "issue:42915", - "right": "issue:43742", + "left": "issue:44987", + "right": "issue:45907", "accept": false, - "reason": "Qwen3Moe FP8 config failure and MobileLLM-125M key error are different model-loading bugs." + "reason": "Loading physical-intelligence/fast on transformers>=5.1.0 is unrelated to list_repo_files not forwarding revision in CI." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "MacOS CLIP bus error and TrainerState token-count tracking are unrelated." + "reason": "MacOS bus error on CLIP model load is unrelated to TrainerState token-count tracking." } ] }, @@ -5784,9 +5703,9 @@ "issue:42175|issue:44488", "issue:45137|issue:45405", "issue:43906|issue:45381", + "issue:44737|issue:45588", "issue:28282|issue:34689", "issue:42915|issue:43749", - "issue:44737|issue:45588", "issue:43742|issue:44291", "issue:43526|issue:45479", "issue:43582|issue:43782", @@ -5801,14 +5720,14 @@ "issue:43643|issue:45375", "issue:43452|issue:44488" ], - "prepared_review_unit_hash": "bdfd4358f8ffcc886a94ba198a347bea5a537bfe82fa965ab865720d1dda2565", + "prepared_review_unit_hash": "27922551fc9691a35b2ee5705e8d84f5adb6ab6923711f46e94d071bcf194b92", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, @@ -5823,210 +5742,198 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly heterogeneous and does not look like one duplicate cluster. I only accept the SigLIP2 pair as likely the same underlying wrong-results bug; the rest are distinct model, tokenizer, config, training, or backend issues.", - "confidence": 0.83, - "canonical_issue_reason": "If one issue must anchor the set, issue 43994 is the broadest user-facing report of the SigLIP2 wrong-results problem and best matches the only plausible duplicate pair.", + "summary": "Most items are unrelated bug reports across different models/subsystems. I found one plausible duplicate pair around SigLIP2; the rest should stay separate.", + "confidence": 0.87, + "canonical_issue_reason": "No single canonical issue for the full set; the only believable duplicate pair is the SigLIP2 reports, where 43994 is the clearer anchor because it captures the user-visible failure and broader impact.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43994 is the best representative issue because it describes the symptom end users see, while 43493 is the more diagnostic counterpart.", + "best_issue_reason": "43994 is the best overall representative among these issues because it describes a broad, user-facing regression in a widely used model path, while most others are narrower or unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Different subsystems: Qwen3Moe FP8 config failure vs AutoImageProcessor URL loading." + "reason": "Different models and failures: Qwen3Moe FP8 loading vs AutoImageProcessor URL loading." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "Unrelated bugs: Qwen3Moe FP8 config vs CLIPTokenizer max_length handling." + "reason": "Unrelated bugs: Qwen3Moe FP8 config error vs CLIPTokenizer max_length behavior." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "Different code paths: DiaConfig save/load token IDs vs MXFP4 quantization fallback on git-oss-20b." + "reason": "Different code paths: config save/load token IDs vs MXFP4 quantization fallback." }, { "left": "issue:43317", "right": "issue:45538", "accept": false, - "reason": "Device-mapping/dequantized offload issue is unrelated to tokenizer max_length defaults." + "reason": "Device-map/offload loading bug is unrelated to CLIPTokenizer max_length." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Missing TensorFlow backend in torch install is unrelated to a missing Qwen2.5-VL class import." + "reason": "Backend packaging issue vs missing Qwen2.5-VL class import." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both involve video, but they target different Qwen model families and different failure modes/code paths." + "reason": "Both are video-related, but they target different Qwen model families and failure modes; not enough evidence of one underlying bug." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Different models and bugs: Qwen3OmniMoe video processing vs Qwen3-vl-embedding StopIteration." + "reason": "Different video-input errors in different models; not the same concrete bug." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Different dtype bugs in different architectures; not the same concrete fix." + "reason": "Both mention dtype, but they affect different architectures and symptoms." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "Tokenizer construction error vs model-specific tokenizer loading regression are not the same underlying bug." + "reason": "Tokenizer errors, but one is AddedToken argument handling and the other is loading a specific tokenizer." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "Backend installation packaging issue is unrelated to Python 3.14 import/load failures." + "reason": "Backend installation mismatch is unrelated to Python 3.14 import/load failure." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "Different vision models and symptoms: SigLIP2 zero probabilities vs Perceiver non-default resolution failure." + "reason": "Different models and failures: SigLIP2 zero probabilities vs Perceiver non-default resolution." }, { "left": "issue:42175", "right": "issue:44488", "accept": false, - "reason": "Packaging/backend omission vs a specific model loading regression; unrelated." + "reason": "Packaging/backend issue vs model loading failure for a specific checkpoint." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 runtime error is unrelated to a PEFT version pin bump." + "reason": "DeepSpeed ZeRO3 runtime error is unrelated to PEFT version bump." }, { "left": "issue:43906", "right": "issue:45381", "accept": false, - "reason": "Generic reproduction of another issue vs a Qwen2.5-VL video-position regression; not the same bug." + "reason": "No clear shared bug: isolated reproduction of an old issue vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU placement bug is unrelated to flash_attention s_aux=None crash." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "PyTorch-missing ImportError and Llama 3.2 Vision load breakage are different failures." + "reason": "Generic PyTorch ImportError vs a specific Llama 3.2 Vision loading regression." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Different failure modes: Qwen3Moe FP8 config vs FSDP CPU RAM efficient loading." - }, - { - "left": "issue:44737", - "right": "issue:45588", - "accept": false, - "reason": "XLNet CPU allocation issue is unrelated to flash-attention sinkless-model crash." + "reason": "Different training/loading problems with no shared failing code path." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "Both are load-time errors, but one is a key error for a specific model and the other is an init_empty_weights argument mismatch." + "reason": "Both are load-time issues, but one is a missing key in MobileLLM and the other is an unexpected init_empty_weights argument." }, { "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to single-label classification zero-loss behavior." + "reason": "Image processor label reduction bug vs sequence-classification zero-loss bug." }, { "left": "issue:43582", "right": "issue:43782", "accept": false, - "reason": "Apple Silicon allocator warmup TypeError and Qwen3VL weight_only load error are different code paths." + "reason": "Apple Silicon allocator TypeError is unrelated to Qwen3VL weight_only loading failure." }, { "left": "issue:43540", "right": "issue:45823", "accept": false, - "reason": "Different models and components: Qwen3OmniMoe video parsing vs Gemma4 device mismatch." + "reason": "Video processing error vs Gemma4 PLE device mismatch; different subsystems." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Both mention RoPE/positioning, but they concern different models and distinct implementation mistakes." + "reason": "Different RoPE-related model bugs with different symptoms and affected models." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "PyTorch missing ImportError is unrelated to LayoutLMv3 box-content validation messaging." + "reason": "PyTorch missing import error is unrelated to LayoutLMv3 box-validation messaging." }, { "left": "issue:43493", "right": "issue:43994", "accept": true, - "reason": "Both point to the same SigLIP2 wrong-results / implementation-mismatch problem; one is the broader symptom report." + "reason": "Both are SigLIP2 regressions describing the same underlying model discrepancy, with 43994 showing the user-facing symptom and 43493 giving the implementation-vs-JAX diagnosis." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token counting and CustomTrainer compute_loss signature mismatch are different trainer issues." + "reason": "TrainerState token counting bug vs CustomTrainer compute_loss signature mismatch." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "AddedToken special-arg duplication is unrelated to Mistral regex patching and backend_tokenizer access." + "reason": "Different tokenizer failures: AddedToken special-arg duplication vs Mistral regex patch crash." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "init_empty_weights argument handling is unrelated to loading AutoImageProcessor from a URL." + "reason": "Model init-empty-weights argument bug is unrelated to AutoImageProcessor URL loading." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 save_pretrained key regression is unrelated to AutoTokenizer producing wrong token IDs across models." + "reason": "save_pretrained visual-key regression is not the same as AutoTokenizer wrong token IDs." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both concern config fields, but one is trust_remote_code loading behavior and the other is a specific Qwen3_5MoeVisionConfig field omission." - }, - { - "left": "issue:43782", - "right": "issue:44479", - "accept": false, - "reason": "Qwen3VL weight_only load error is unrelated to the v5.3.0 video-input regression across Qwen vision models." - }, - { - "left": "issue:44938", - "right": "issue:45920", - "accept": false, - "reason": "Python 3.14 compatibility/load failure is unrelated to tokenizer ID regression across multiple models." + "reason": "Both involve config fields, but one is trust_remote_code AutoConfig loss and the other is a specific Qwen vision config field." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "GGUF loading breakage and a specific model/tokenizer load failure are not the same underlying bug." + "reason": "GGUF tokenizer/model loading breakage is unrelated to the sleng-bert loading report." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded overall: the packet supports treating 43994 as the user-facing anchor and 43493 as the more diagnostic SigLIP2 counterpart. The rest of the soft-edge rejections are conservative and aligned with the issue titles; no pair is overstated as a duplicate cluster beyond the one accepted SigLIP2 match." + "feedback": "The summary is grounded in the packet and stays conservative. It correctly treats most pairs as unrelated and only tentatively groups the SigLIP2 issues, which is the one pair whose titles strongly align. I don\u2019t see any overstatement that would require rejection." }, "evaluator_used": true, "retried": false, @@ -6042,15 +5949,16 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ "issue:24643", "issue:29127", + "issue:30064", + "issue:33666", "issue:38175", - "issue:41720", "issue:41762", "issue:43295", "issue:43452", @@ -6064,7 +5972,6 @@ "issue:43742", "issue:43749", "issue:43819", - "issue:43906", "issue:44361", "issue:44442", "issue:44464", @@ -6123,204 +6030,204 @@ "issue:43643|issue:44561", "issue:43653|issue:44361", "issue:44361|issue:44442", - "issue:41720|issue:43906" + "issue:30064|issue:33666" ], - "prepared_review_unit_hash": "fc02e1b0dc39bf51e9720416aa552e40030e2c8d196cf1e35a6827557f7287e4", + "prepared_review_unit_hash": "ed851354ef47701a33e732a66e00c4f0b7509399f5561e92cac6613d8ff85ae2", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 36062, - "estimated_input_tokens": 9016, - "estimated_eval_tokens": 18288 + "serialized_chars": 35983, + "estimated_input_tokens": 8996, + "estimated_eval_tokens": 18248 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the soft pairs cover many unrelated Transformers bugs, regressions, and model-specific loading failures. The only plausible duplicate-like pair is the v5 tied-weights/save_pretrained serialization regression.", - "confidence": 0.41, - "canonical_issue_reason": "issue:44466 is the clearest concrete root-cause report: it describes device-dependent incorrect serialization of tied `lm_head.weight` in v5, which is the most actionable and likely umbrella bug among the close matches.", + "summary": "Most items are unrelated single-issue reports. The only plausible duplicate/near-duplicate pair is the tied-weights save/serialization regression in v5. No PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "issue:44466 is the broadest, most concrete report in the only likely duplicate family: it describes the tied-weights serialization problem in v5 in a way that can absorb model-specific variants.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44466 is the strongest representative issue because it is specific, regression-oriented, and directly tied to a likely fixable code path; its scope also plausibly subsumes the broader tied-weights complaint in issue:45005.", + "best_issue_reason": "issue:44466 is the best representative issue for triage because it focuses on the underlying save_pretrained/tied-weights regression rather than a narrow model instance.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Both are multimodal loading failures, but they involve different models and symptoms, so they do not look like the same bug." + "reason": "Different failures: a janus image-generation test case versus loading Phi-4 multimodal models. No shared code path." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "One is an image-processor dependency regression; the other is a Mistral tokenizer regex crash. Different code paths." + "reason": "Unrelated subsystems: PIL image processor torchvision dependency vs Mistral regex/tokenizer loading crash." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "Different loading failures with unrelated error modes: a key error vs. a float8 storage lookup TypeError." + "reason": "Different model/load-time errors: MobileLLM key error versus Float8 storage lookup failure." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding-window logic and AutoImageProcessor URL loading are unrelated." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to loading AutoImageProcessor from a URL." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "One is a PEFT version pin problem; the other is stale auto_mappings entries breaking CI. Not the same bug." + "reason": "One is a dependency version bump issue; the other is stale auto_mappings references to removed configs." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Both are regressions, but one is RTDetrV2 output drift and the other is Qwen3.5 save_pretrained corruption. Different concrete issues." + "reason": "Different bugs: RTDetrV2 output changes versus Qwen3.5 save_pretrained regression." }, { "left": "issue:43688", "right": "issue:43749", "accept": false, - "reason": "Aux-loss normalization and FSDP CPU RAM-efficient loading are different training/loading problems." + "reason": "Both involve training/runtime infrastructure, but one is auxiliary-loss normalization and the other is FSDP loading." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Both concern Qwen2.5-VL, but one is chat-template batching and the other is video vision-position IDs. Different bugs." + "reason": "Different chat-template/video-position-id problems; not the same concrete bug." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "BitNet packed-weight loading and Perceiver resolution handling are unrelated." + "reason": "BitNet accelerated-loading unpacking bug versus Perceiver image classification interpolation failure." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "A save_pretrained shard-size question and an AutoProcessor kwargs forwarding bug are not duplicates." + "reason": "save_pretrained shard size question is unrelated to AutoProcessor kwargs forwarding." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "Gemma3 DeepSpeed ZeRO-3 loading failure and trainer.evaluate after train are different failure modes." + "reason": "Gemma3 DeepSpeed ZeRO-3 load failure is not the same as trainer.evaluate() failing after train()." }, { "left": "issue:24643", "right": "issue:29127", "accept": false, - "reason": "DeepSpeed weight-dimension runtime error vs. LayoutLMv3 error-message clarity; same area of stack traces, but not the same underlying bug." + "reason": "DeepSpeed weight-dimension runtime error is unrelated to LayoutLMv3 box validation messaging." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation and granite_speech strict-config loading are distinct config bugs." + "reason": "Qwen3.5 num_labels propagation is unrelated to granite_speech strict-config loading." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both involve apply_chat_template, but the failing inputs and failure mode differ enough that they do not look like one bug." + "reason": "Both are apply_chat_template-related, but they fail on different inputs and code paths." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "gguf_file loading breakage and BLIP-2 dtype propagation are unrelated." + "reason": "Tokenizer/model loading via gguf_file is unrelated to Blip2 dtype propagation." }, { "left": "issue:43452", "right": "issue:44843", "accept": false, - "reason": "Both involve loading/tokenizer paths, but one is gguf_file support and the other is offline Mistral regex patching." + "reason": "Different tokenizer-loading failures: gguf_file support versus offline model_info lookup in mistral regex patching." }, { "left": "issue:44466", "right": "issue:45005", "accept": true, - "reason": "Both point to the same tied-weights/save_pretrained regression in v5, with device-dependent serialization behavior affecting translation models as a manifestation." + "reason": "Both report the same tied-weights regression in v5 save/serialization behavior, with 45005 appearing model-specific and 44466 describing the underlying device-dependent bug." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "DAC latent reconstruction and Whisper batch_decode special-token handling are separate behaviors." + "reason": "DAC latent conversion mismatch and Whisper batch_decode special-token handling are unrelated." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer/processor regressions, but one is processor.tokenizer access and the other is Kimi-K2.5 codec handling; not the same fix." + "reason": "Both mention tokenizer/processor regressions, but one is custom processor.tokenizer access while the other is Kimi codec handling and warning behavior." }, { "left": "issue:44936", "right": "issue:45823", "accept": false, - "reason": "trainer.evaluate after train and Gemma4 PLE device mismatch are different runtime problems." + "reason": "Trainer lifecycle failure is unrelated to Gemma4 PLE device-map forward mismatch." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer special-token issues, but BigBird mask-token registration and mdeberta BOS/EOS behavior are different concrete bugs." + "reason": "BigBird special-token registration bug is unrelated to mdeberta-v3 BOS/EOS addition regression." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency and recurrent-state reset with cache are distinct generation/caching bugs." + "reason": "Different generation bugs: compiled forward chunking versus Qwen3.5 recurrent state reset with cache." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "SigLIP2 zero probabilities and processor.tokenizer regression are unrelated." + "reason": "Siglip2 zero-probabilities issue is unrelated to processor.tokenizer regression." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "Both mention trust_remote_code, but one is missing fields from AutoConfig and the other is an is_torch_fx_available removal regression." + "reason": "Both involve trust_remote_code, but one is missing AutoConfig fields and the other is v5 removal of is_torch_fx_available." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "Different tokenizer failures on different models; no shared concrete bug." + "reason": "Different tokenizer failures: BigBird mask-token registration versus MLukeTokenizer AttributeError." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "MLuke tokenizer task error and AutoTokenizer failing for FastSpeech2Conformer are separate loading/tokenizer problems." + "reason": "MLukeTokenizer task AttributeError is unrelated to FastSpeech2ConformerTokenizer loading." }, { - "left": "issue:41720", - "right": "issue:43906", + "left": "issue:30064", + "right": "issue:33666", "accept": false, - "reason": "Qwen3 auto device mapping cuda assert and an isolated reproduction of another issue are not the same underlying bug." + "reason": "Void segmentation map processing and Qwen2-VL multi-GPU training are different problems." } ] }, "evaluator_result": { "accept": true, - "feedback": "The assessment is conservative and grounded in the packet. The summary does not overclaim duplication beyond the closely related tied-weights/save_pretrained pair, and the other soft-edge rejections are supported by the issue titles and described failure modes. The accepted 44466/45005 pair is a plausible match rather than an overstated certainty." + "feedback": "The summary is grounded in the packet, and the only accepted soft-edge pair is a conservative, plausible tied-weights duplicate family. The other verdicts stay appropriately cautious and do not overstate shared code paths or bug identity." }, "evaluator_used": true, "retried": false, @@ -6336,13 +6243,12 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ "issue:28282", - "issue:30064", "issue:32090", "issue:33357", "issue:33666", @@ -6355,8 +6261,10 @@ "issue:43452", "issue:43493", "issue:43504", + "issue:43525", "issue:43531", "issue:43550", + "issue:43572", "issue:43577", "issue:43606", "issue:43646", @@ -6390,23 +6298,23 @@ "issue:45812" ], "soft_pairs": [ + "issue:41720|issue:43906", "issue:43688|issue:45440", - "issue:30064|issue:33666", "issue:43688|issue:45704", "issue:43577|issue:43697", "issue:28282|issue:32090", "issue:28282|issue:33357", + "issue:28282|issue:35141", "issue:43646|issue:43749", "issue:44843|issue:44991", - "issue:28282|issue:35141", "issue:43688|issue:45701", "issue:43653|issue:43927", "issue:43531|issue:43742", "issue:43550|issue:43927", "issue:44464|issue:44977", "issue:44479|issue:44625", - "issue:43577|issue:44554", "issue:43550|issue:44361", + "issue:43577|issue:44554", "issue:43577|issue:45588", "issue:43122|issue:43906", "issue:43504|issue:43550", @@ -6421,242 +6329,249 @@ "issue:43278|issue:43688", "issue:43493|issue:43825", "issue:33666|issue:36010", - "issue:44568|issue:45812", "issue:43606|issue:43927", - "issue:43720|issue:43927" + "issue:43720|issue:43927", + "issue:44568|issue:45812", + "issue:43525|issue:43572" ], - "prepared_review_unit_hash": "47aeb53ed2c6ba06f66e593266cb9674e0541eb311ec30778d870df6043397e3", + "prepared_review_unit_hash": "2b5359add63ec35c01e6822a4d9bf03bbcedf01e415c3393aeb2b1ad78d477c4", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 34, - "serialized_chars": 37287, - "estimated_input_tokens": 9322, - "estimated_eval_tokens": 18900 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38168, + "estimated_input_tokens": 9542, + "estimated_eval_tokens": 19340 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous, but a smaller theme around tokenizer regressions/loading behavior in v5 stands out. I did not find any soft pair that is clearly the same concrete bug or change, so all soft edges are rejected.", - "confidence": 0.38, - "canonical_issue_reason": "issue:45701 is the broadest open tokenization-regression umbrella and is the best place to attach version-induced tokenizer behavior duplicates.", + "summary": "This is a heterogeneous cluster spanning unrelated tokenizer, model-loading, training, generation, and config bugs. None of the proposed soft pairs look like the same underlying bug/change, so all soft edges are rejected.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the most reusable triage target for tokenizer-change reports across versions, even though the cluster overall is mixed.", + "best_issue_reason": "issue:45812 is the strongest standalone representative: it is a concrete, reproducible tokenizer regression with broad impact across Granite models and clear v4\u2192v5 behavior change.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43688", - "right": "issue:45440", + "left": "issue:41720", + "right": "issue:43906", "accept": false, - "reason": "Different models and failures: auxiliary-loss normalization vs implementation divergence in DeepSeekV3MoE." + "reason": "Different failures: auto device mapping CUDA assert on A800 vs an isolated reproduction of another issue." }, { - "left": "issue:30064", - "right": "issue:33666", + "left": "issue:43688", + "right": "issue:45440", "accept": false, - "reason": "Image segmentation map processing is unrelated to Qwen2-VL multi-GPU training." + "reason": "Different bugs and code paths: auxiliary-loss normalization vs native DeepSeekV3 divergence." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "Aux-loss normalization bug is unrelated to T5/apex RMSNorm memory leak." + "reason": "Different model/code paths: expert-loss normalization vs T5 apex FusedRMSNorm memory leak." }, { "left": "issue:43577", "right": "issue:43697", "accept": false, - "reason": "BLIP2 dtype handling and RTDetrV2 output differences are distinct bugs." + "reason": "Different symptoms and models: Blip2 dtype remaining fp32 vs RTDetrV2 output differences." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "ImportError for missing PyTorch and Trainer broadcast TypeError are unrelated." + "reason": "ImportError for missing PyTorch vs Trainer GPU broadcast NoneType TypeError are unrelated." }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "Missing PyTorch ImportError and MacOS CLIP bus error are unrelated." + "reason": "Missing PyTorch ImportError vs MacOS CLIP bus error are unrelated failures." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "Different bug classes: dependency import failure vs embedding resize reinitialization." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "Custom model initialization breakage and FSDP CPU RAM loading are different code paths." + "reason": "Both are v5/loading-related, but they target different mechanisms: custom init breakage vs FSDP CPU RAM efficient loading." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Offline Mistral regex patch issue and tokenizer loading failure for est-roberta are not the same bug." - }, - { - "left": "issue:28282", - "right": "issue:35141", - "accept": false, - "reason": "Environment ImportError and embedding-resize reinit bug are unrelated." + "reason": "Different tokenizer paths: unconditional model_info call in Mistral regex patch vs tokenizer loading for EMBEDDIA/est-roberta." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Auxiliary-loss normalization does not match a tokenizer regression." + "reason": "Auxiliary-loss normalization and tokenization changes are unrelated problems." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "BigBird mask-token decode issue and DiaConfig custom-ID loss are different tokenizer/config bugs." + "reason": "BigBirdTokenizer special-token registration vs DiaConfig save/load token-ID loss are different bugs." }, { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Qwen3-MoE sliding_window bug and MobileLLM load key error are unrelated." + "reason": "Qwen3-MoE sliding_window issue and MobileLLM load key error are different code paths." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "torch.compile/SDPA failure and DiaConfig save-load generation error are unrelated." + "reason": "torch.compile/SDPA behavior in Bamba is unrelated to DiaConfig token-ID persistence." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is different from Qwen3.5 flash-attention generation failure." + "reason": "Chunked generation with compiled forward vs Qwen3.5 flash-attention generation failure are different issues." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Video-input regression and num_labels propagation are unrelated Qwen3.5 issues." + "reason": "Video-input regression for Qwen VL models vs num_labels propagation in Qwen3.5 are unrelated." }, { - "left": "issue:43577", - "right": "issue:44554", + "left": "issue:43550", + "right": "issue:44361", "accept": false, - "reason": "BLIP2 dtype loading and MPS attention correctness are unrelated." + "reason": "Bamba compile/SDPA bug vs MLukeTokenizer AttributeError are unrelated." }, { - "left": "issue:43550", - "right": "issue:44361", + "left": "issue:43577", + "right": "issue:44554", "accept": false, - "reason": "Bamba torch.compile SDPA failure and MLukeTokenizer AttributeError are unrelated." + "reason": "Blip2 dtype handling and an MPS attention correctness bug are different problems." }, { "left": "issue:43577", "right": "issue:45588", "accept": false, - "reason": "BLIP2 dtype handling is unrelated to flash_attention.py crashing on s_aux=None." + "reason": "Blip2 dtype retention vs flash_attention.py crashing on s_aux=None are unrelated." }, { "left": "issue:43122", "right": "issue:43906", "accept": false, - "reason": "Both concern tokenization, but one is a version-to-version difference and the other is a reproduction of a separate referenced issue; not enough to call the same bug." + "reason": "Different tokenization regression reports with no clear shared code path or concrete duplicate evidence." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "Legacy-field BEiT loading and Bamba torch.compile SDPA are different problems." + "reason": "Beit legacy-field loading failure vs Bamba torch.compile/SDPA bug are unrelated." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Legacy-field BEiT loading and BitNet packed-weight loading are unrelated." + "reason": "Beit preset loading and BitNet packed-weight unpacking are different load-time bugs." }, { "left": "issue:32090", "right": "issue:34567", "accept": false, - "reason": "Trainer broadcast TypeError and num_input_tokens_seen not updating are unrelated trainer issues." + "reason": "Trainer broadcast NoneType error vs num_input_tokens_seen not updating are separate Trainer issues." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Qwen3 auto device mapping cuda assert and compiled-forward chunked generation are different failures." + "reason": "CUDA assert during auto device mapping is unrelated to chunked generation inconsistency." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "BEiT legacy-field loading and OmDet-Turbo processor size mismatch are unrelated." + "reason": "Legacy config field loading vs processor/model input-size mismatch are different bugs." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer broadcast TypeError is unrelated to Qwen2-VL multi-GPU training." + "reason": "Trainer broadcast NoneType error and Qwen2-VL multi-GPU training are not the same failure." }, { "left": "issue:43452", "right": "issue:44451", "accept": false, - "reason": "gguf_file/from_pretrained breakage and ScandiBERT load failure are too vague to be the same concrete bug." + "reason": "GGUF/from_pretrained breakage and a specific model-load failure for ScandiBERT are different issues." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "Qwen3.5 output_hidden_states bug and save_pretrained visual-key regression are different code paths." + "reason": "Qwen3.5 hidden-states bug vs save_pretrained visual-encoder key regression are unrelated." }, { "left": "issue:43742", "right": "issue:44821", "accept": false, - "reason": "MobileLLM key error and AutoImageProcessor-from-URL failure are unrelated loading issues." + "reason": "MobileLLM key error vs AutoImageProcessor URL loading failure are different loading paths." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "Embedding dtype mismatch in evaluate and auxiliary-loss normalization are unrelated." + "reason": "Embedding dtype drift in evaluate vs auxiliary-loss normalization are unrelated." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 implementation discrepancy and pipeline translation-message bug are unrelated." + "reason": "SigLIP2 implementation discrepancy vs pipeline translation error-message regression are different concerns." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Qwen2-VL multi-GPU training and GenerationMixin import error are unrelated." - }, - { - "left": "issue:44568", - "right": "issue:45812", - "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and different behaviors; not the same concrete bug." + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure do not share the same bug." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch and DiaConfig token-ID persistence are unrelated." + "reason": "CPU-offload device mismatch for bark-small vs DiaConfig token-ID loss are unrelated." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "BitNet packed-weight loading and DiaConfig save/load token-ID loss are unrelated." + "reason": "BitNet accelerate-loading unpacking bug vs DiaConfig token-ID persistence bug are different." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and code paths: special-token injection for mdeberta-v3 vs wrong token IDs for Granite." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config and missing pad_token_idx in StableLmConfig are similar in shape but not the same concrete bug or fix." } ] }, @@ -6675,8 +6590,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -6693,9 +6608,9 @@ "issue:43334", "issue:43493", "issue:43504", - "issue:43525", "issue:43540", "issue:43572", + "issue:43575", "issue:43606", "issue:43646", "issue:43653", @@ -6730,9 +6645,8 @@ "issue:45823" ], "soft_pairs": [ - "issue:43525|issue:43572", - "issue:43756|issue:44913", "issue:43262|issue:43295", + "issue:43756|issue:44913", "issue:43720|issue:44361", "issue:28218|issue:33357", "issue:33666|issue:36331", @@ -6761,228 +6675,229 @@ "issue:43504|issue:43653", "issue:44464|issue:45584", "issue:43606|issue:44112", - "issue:43994|issue:44493" + "issue:43994|issue:44493", + "issue:43575|issue:45405" ], - "prepared_review_unit_hash": "429c7702711d6fa69a801de64a83ddf3caf14b176b2badba66a605455806c428", + "prepared_review_unit_hash": "9fff9d47fd7ccaa3e4cfbc8ef9993c69c2f175e6b0b20c345c2c816c5b112857", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37565, - "estimated_input_tokens": 9392, - "estimated_eval_tokens": 19040 + "serialized_chars": 37543, + "estimated_input_tokens": 9386, + "estimated_eval_tokens": 19028 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is a grab-bag of mostly unrelated transformers issues. All soft pairs look like false duplicates: they touch different models, APIs, failure modes, or code paths, even when they share broad themes like configs, tokenizers, loading, or multimodal support.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue exists because the items are not duplicates of one another. If a representative issue is needed, #43295 is the clearest standalone regression report, but it does not subsume the others.", + "summary": "The cluster is heterogeneous and the soft pairs are only broad thematic matches, not true duplicates. I would reject all proposed duplicate edges and avoid selecting a canonical issue/PR for the cluster.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "#43295 is the strongest standalone issue to represent a bug report: it is a concrete regression with a specific failure mode and actionable reproduction context.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43525", - "right": "issue:43572", + "left": "issue:43262", + "right": "issue:43295", "accept": false, - "reason": "Both mention missing pad token fields, but they affect different config classes (Llama4 vs StableLm) and different loading paths." + "reason": "Audio sample-rate default bug vs processor/tokenizer regression; different subsystems and failure modes." }, { "left": "issue:43756", "right": "issue:44913", "accept": false, - "reason": "Both involve rotary/rope config behavior, but one is a Smollm3 architecture mismatch and the other is a GPTNeoX reload/default persistence bug." - }, - { - "left": "issue:43262", - "right": "issue:43295", - "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and processor.tokenizer/image handling are different APIs and different failure modes." + "reason": "Both mention rotary settings, but one is a Smollm3 RoPE-layer implementation issue and the other is a GPTNeoX config reload bug." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "BitNet packed-weight loading and MLukeTokenizer task errors are unrelated subsystems with unrelated symptoms." + "reason": "BitNet packed-weight loading failure vs MLukeTokenizer task-time AttributeError; unrelated code paths." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer spacing after added tokens is unrelated to a MacOS bus error when loading a CLIP model." + "reason": "Tokenizer spacing bug vs MacOS bus error on a CLIP model; not the same defect." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "Multi-GPU Qwen2-VL training and a custom compute_loss signature break are different training issues, not the same bug." + "reason": "Qwen2-VL multi-GPU training request vs custom trainer loss-signature TypeError; different problems." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Both concern SigLIP2, but one is a docs/example mismatch and quantization failure while the other is implementation parity with JAX." + "reason": "Documentation/example errors vs SigLIP2 implementation parity discrepancy; not the same underlying bug." }, { "left": "issue:43540", "right": "issue:44977", "accept": false, - "reason": "Qwen3OmniMoe video-input validation and Qwen3.5 flash-attention generation are separate code paths and failures." + "reason": "Qwen3OmniMoe video-input processing vs Qwen3.5 flash-attention generation issue; different code paths." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 parity issues and ZeRO-3 gradient growth are unrelated bugs." + "reason": "SigLIP2 model mismatch issue vs DeepSpeed/gradient behavior regression; unrelated." }, { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "A model-loading regression for Phi-4 multimodal is unrelated to a PEFT version pin being ahead of PyPI." + "reason": "Phi-4 multimodal loading failure vs PEFT version pin bump; no shared bug." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "Embedding dtype drift in evaluate and abnormal gradients under ZeRO-3 are distinct symptoms with different likely causes." + "reason": "Eval/train dtype mismatch vs ZeRO-3 gradient inflation; different defects." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Both mention missing padding-related config fields, but they are different models and different missing attributes." + "reason": "Both involve missing pad-token fields, but in different models/config layers and not the same concrete bug." }, { "left": "issue:44625", "right": "issue:45584", "accept": false, - "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription generation failures are unrelated." + "reason": "Qwen3.5 num_labels propagation bug vs Whisper empty-transcription generation failure; unrelated." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Both are Qwen3 MOE loading reports, but one is accelerate+deepspeed weight conversion and the other is a dev-version model init break; not the same concrete bug." + "reason": "Both are Qwen MOE-related, but one is accelerate/deepspeed weight conversion and the other is multimodal loading; not mergeable as one fix." }, { "left": "issue:43756", "right": "issue:45083", "accept": false, - "reason": "Smollm3 RoPE layer count mismatch is unrelated to a helper function behavior bug in qwen3_omni_moe." + "reason": "Smollm3 RoPE-layer mismatch vs qwen3_omni_moe feature-length helper behavior; different bugs." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "Both touch Qwen multimodal config handling, but one is a missing pad_token_id attribute and the other is from_config not unwrapping text_config; different fixes." + "reason": "Missing pad_token_id attribute vs unwrap text_config in from_config; related family, but distinct code paths." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "Loading a specific model and offline-mode tokenizer patch behavior are unrelated." + "reason": "Model load failure for a specific model vs offline-mode model_info call; unrelated." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Gemma4 text-only mm_token_type_ids defaults and Gemma4 multi-GPU inference OOM are different bugs with different causes." + "reason": "Text-only token-type defaulting vs multi-GPU inference OOM; different issues." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "Missing Qwen2.5 VL import and Float8 storage lookup failure are unrelated compatibility errors." + "reason": "ImportError for a Qwen2.5-VL class vs missing Float8 storage type; unrelated failures." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model initialization in v5 and lm_head serialization depending on device are separate model-core regressions." + "reason": "Custom model initialization regression vs device-dependent serialization inconsistency; not the same bug." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "A SigLIP2 docs/example issue is not the same as a regression in processor.tokenizer/image handling." + "reason": "SigLIP2 docs/example problems vs processor/tokenizer regression; different scopes and code paths." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Both are Gemma4-related, but one is inference OOM while the other is a PLE device mismatch during forward." + "reason": "Both Gemma4-related, but one is CUDA OOM and the other is a device mismatch during forward; not the same defect." }, { "left": "issue:43504", "right": "issue:43606", "accept": false, - "reason": "BEiT pretrained loading with a legacy field and Bark CPU-offload device mismatch are unrelated." + "reason": "Beit legacy-field loading bug vs Bark CPU-offload device mismatch; unrelated." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Float8 storage type resolution and flash_attn import metadata on Python 3.13 are unrelated compatibility issues." + "reason": "Float8 storage lookup failure vs flash_attn import-utils KeyError on Python 3.13; different issues." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing StableLm pad_token_idx and init_empty_weights receiving an unexpected _is_hf_initialized argument are different constructor/config regressions." + "reason": "Both are v5 init/config regressions, but one is a StableLmConfig field issue and the other is an init_empty_weights argument change." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "A MacOS CLIP bus error and Qwen2-VL multi-GPU training are completely different failures." + "reason": "MacOS CLIP bus error vs Qwen2-VL multi-GPU training; no common bug." }, { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Continuous-batching multimodal crash and KV-cache continuation indexing bug are both serve-related, but they are different concrete failures." + "reason": "Serve crash on multimodal model input conversion vs KV-cache continuation tensor indexing; different serve paths." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "DiaConfig save/load token-ID loss and MLukeTokenizer task AttributeError are unrelated model/tokenizer bugs." + "reason": "DiaConfig save/load token-ID loss vs MLukeTokenizer AttributeError; unrelated." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "BEiT semantic-segmentation preset loading and BigBirdTokenizer mask-token registration are different components and symptoms." + "reason": "Legacy-field model load bug vs BigBirdTokenizer special-token registration issue; different fixes." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency and Whisper empty-transcription handling are distinct generation bugs." + "reason": "Chunked generation inconsistency under compiled forward vs Whisper empty transcription after align_special_tokens; unrelated." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "A real device-mismatch bug and a stale CI test for GraniteSpeech are not the same underlying issue." + "reason": "Real CPU-offload device mismatch bug vs a stale GraniteSpeech CI test; not duplicates." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 nonsensical outputs with AutoModel/pipeline and unexpected position-id keys across many models are unrelated." + "reason": "SigLIP2 pipeline output issue vs broad unexpected position-id keys; not the same underlying problem." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM on a specific model vs unreleased PEFT version bump; unrelated." } ] }, @@ -7001,8 +6916,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -7020,7 +6935,6 @@ "issue:43493", "issue:43540", "issue:43572", - "issue:43575", "issue:43582", "issue:43696", "issue:43701", @@ -7037,6 +6951,7 @@ "issue:44792", "issue:44849", "issue:44898", + "issue:44936", "issue:44964", "issue:44977", "issue:44987", @@ -7056,7 +6971,6 @@ "issue:45823" ], "soft_pairs": [ - "issue:43575|issue:45405", "issue:43278|issue:43701", "issue:44792|issue:45405", "issue:44521|issue:45290", @@ -7084,210 +6998,211 @@ "issue:43572|issue:44987", "issue:45200|issue:45325", "issue:37428|issue:43824", - "issue:43572|issue:43957" + "issue:43572|issue:43957", + "issue:43844|issue:44936" ], - "prepared_review_unit_hash": "da1ad87c879a85feb47c6ef96a2dec2b60638c463f37f1386f1e0aba7c0c9639", + "prepared_review_unit_hash": "eee1573adc68cd615b64d0c7f0f40d98a8056232111f73de13044540dfe69725", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36424, - "estimated_input_tokens": 9106, - "estimated_eval_tokens": 18468 + "serialized_chars": 36425, + "estimated_input_tokens": 9107, + "estimated_eval_tokens": 18470 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft-edge pair looks like the same underlying bug/change. The items span unrelated subsystems and different failure modes, so none should be merged as duplicates.", - "confidence": 0.95, + "summary": "This cluster is heterogeneous: the soft links mostly share broad areas like tokenization, generation, or multimodal models, but they do not describe the same concrete bug or change. No duplicate canonical issue/PR stands out.", + "confidence": 0.94, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43575", - "right": "issue:45405", - "accept": false, - "reason": "Qwen2-57B tensor-parallel OOM and a PEFT minimum-version bump are unrelated." - }, { "left": "issue:43278", "right": "issue:43701", "accept": false, - "reason": "BF16-to-FP32 dtype drift in evaluate is a different bug from checkpoint key mismatches on resume." + "reason": "BF16->FP32 dtype drift in evaluation and resume_from_checkpoint key mismatch are different failures with different code paths." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "A Janus model test failure is unrelated to the PEFT version-release issue." + "reason": "A Janus model test failure and an unreleased PEFT version bump are unrelated." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both involve chat templating, but one is wrong multimodal assistant masks and the other is a crash on empty tool-call assistant messages." + "reason": "Both involve chat templating, but one is wrong assistant masks for multimodal inputs and the other is a crash on tool-call assistant messages; different bugs." }, { "left": "issue:29942", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 test failures and MLFlow not reporting failed jobs are different subsystems and symptoms." + "reason": "Flash Attention 2 test failures and MLflow reporting after job failure are unrelated." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "A local-path import warning is unrelated to a GPT-oss CUDA out-of-memory error." + "reason": "Local-path import warnings and CUDA OOM loading GPT-oss-20b are not the same issue." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient blow-up is a training stability bug, not the same as a Phi-4 model-loading failure." + "reason": "ZeRO-3 gradient growth and a model load/import failure are different problems." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Tokenizer spacing after added tokens is unrelated to TrainerState token-count tracking." + "reason": "Tokenizer spacing after added tokens and TrainerState token counters not updating are separate behaviors." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "An MPS attention correctness issue does not match a PyTorch version check bug in flex_attention imports." + "reason": "An MPS attention correctness bug and a PyTorch version check for AuxRequest import are unrelated." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Non-template node compilation and a flash_attn import KeyError are separate code paths." + "reason": "Template-node compilation errors and a flash_attn import KeyError on Python 3.13 are distinct import/compile paths." }, { "left": "issue:43582", "right": "issue:44977", "accept": false, - "reason": "Apple Silicon allocator TypeError is unrelated to Qwen3.5 flash-attention generation problems." + "reason": "Apple Silicon warmup TypeError and Qwen3.5 flash-attention generation issues are not the same bug." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Video-input processing failure in Qwen3OmniMoe is not the same as Gemma4 mm_token_type_ids defaulting." + "reason": "Video-processing ValueError in Qwen3OmniMoe and Gemma4 text-only mm_token_type_ids defaults are different multimodal issues." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero-probability outputs are unrelated bugs." + "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero probabilities are unrelated model bugs." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "sync_gpus generation-kwargs updating and a Qwen3.5 hidden-states bug affect different generation paths." + "reason": "Generation kwargs after sync_gpus and a Qwen3.5 hidden-states bug do not share the same underlying defect." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "bfloat16 dtype mismatches in inference are unrelated to Wav2Vec2 save/tokenization failures." + "reason": "bfloat16 dtype mismatches in inference and wav2vec2 save/tokenization failures are different code paths." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "The flash-attention helper import error and meta-device loading breakage are different regressions." + "reason": "Missing flash_supports_window_size import and meta-device loading regressions are separate import/model-loading bugs." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Gemma4 mm_token_type_ids defaults and a device_map='auto' device mismatch are separate issues." + "reason": "Gemma4 mm_token_type_ids defaults and device-map forward mismatches are not the same problem." }, { "left": "issue:44315", "right": "issue:44849", "accept": false, - "reason": "Liger Kernel not applied under model_init is unrelated to the Qwen3.5 output_hidden_states bug." + "reason": "Liger Kernel application with model_init and Qwen3.5 hidden-state behavior are unrelated." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS flex-attention correctness and Kimi tokenizer codec handling are different problems." + "reason": "MPS attention correctness and Kimi-K2.5 tokenizer regression are different subsystems and failures." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 HF-vs-JAX discrepancy does not share a code path with resume checkpoint key mismatches." + "reason": "SigLIP2 HF-vs-JAX discrepancy and checkpoint resume key mismatch are unrelated." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Processor/tokenizer regression with multimodal images is unrelated to OmDet-Turbo's 640px vs 224px size mismatch." + "reason": "Processor/tokenizer regression and OmDet-Turbo input-size mismatch are different bugs." }, { "left": "issue:43873", "right": "issue:44464", "accept": false, - "reason": "Quantization offloading failure and compiled-forward chunked-generation inconsistency are distinct bugs." + "reason": "Quantization offloading behavior and chunked generation under compiled forward are not the same issue." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Perceiver resolution handling and RT-DETR memory release are unrelated." + "reason": "Perceiver resolution interpolation failure and RT-DETR memory release leak are unrelated." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Device-dependent lm_head serialization and LoRA merge collapse with extended vocab are different tied-weight issues." + "reason": "lm_head.weight serialization inconsistency and LoRA merge collapse with extended vocab are different failure modes." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "A janus image-generation test failure is not the same as a stale num_return_sequences warning." + "reason": "A Janus model test failure and a stale warning in paged generate() are not the same bug." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is unrelated to loading physical-intelligence/fast on transformers>=5.1.0." + "reason": "Missing pad_token_idx in config and a transformers>=5.1.0 load failure are only loosely related versioned-model issues, not one concrete defect." }, { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Gemma4 mm_token_type_ids defaults and Qwen2.5-VL rope index scaling are separate model bugs." + "reason": "Gemma4 token-type defaults and Qwen2.5-VL rope index scaling are different multimodal bugs." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "A flash_supports_window_size import error is unrelated to the Qwen2_5_VL class import error." + "reason": "These are both import errors, but they concern different missing symbols for different models." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "StableLmConfig pad_token_idx regression and meta-device loading failures are different issues." + "reason": "A missing config field and a meta-device loading regression are not the same underlying change." + }, + { + "left": "issue:43844", + "right": "issue:44936", + "accept": false, + "reason": "ZeRO-3 gradient instability and trainer.evaluate() failing after train() are different training problems." } ] }, @@ -7306,8 +7221,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -7336,7 +7251,6 @@ "issue:43824", "issue:43825", "issue:43827", - "issue:43844", "issue:43856", "issue:43873", "issue:43901", @@ -7350,7 +7264,6 @@ "issue:44610", "issue:44743", "issue:44857", - "issue:44936", "issue:45083", "issue:45127", "issue:45198", @@ -7358,15 +7271,15 @@ "issue:45290", "issue:45381", "issue:45412", - "issue:45584" + "issue:45584", + "issue:45593" ], "soft_pairs": [ - "issue:43844|issue:44936", "issue:43295|issue:43450", "issue:43334|issue:43525", "issue:43824|issue:43975", - "issue:43388|issue:43526", "issue:43295|issue:43873", + "issue:43388|issue:43526", "issue:38175|issue:45412", "issue:42915|issue:43824", "issue:44514|issue:44521", @@ -7374,232 +7287,236 @@ "issue:44442|issue:45198", "issue:28218|issue:35141", "issue:43278|issue:43856", - "issue:43653|issue:44112", "issue:42490|issue:43493", + "issue:43653|issue:44112", "issue:43827|issue:43901", - "issue:44206|issue:45584", "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", "issue:43278|issue:43825", - "issue:45290|issue:45381", - "issue:43606|issue:45083", "issue:37428|issue:43425", + "issue:43606|issue:45083", + "issue:45290|issue:45381", "issue:43335|issue:43550", - "issue:44448|issue:44743" + "issue:44448|issue:44743", + "issue:44857|issue:45593" ], - "prepared_review_unit_hash": "34ec6b70d3e06b07456fbfb13b3fa5779fccb760ae634de10a60afe8e7c0cc10", + "prepared_review_unit_hash": "dd08ec80d0a6a3130b7b16dd0883e6c98f47312e92c9838e41a27d80e6178b7d", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 29, - "serialized_chars": 36574, - "estimated_input_tokens": 9144, - "estimated_eval_tokens": 18544 + "serialized_chars": 35995, + "estimated_input_tokens": 8999, + "estimated_eval_tokens": 18254 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is not a true duplicate cluster; it mixes unrelated tokenizer, processor, model, training, and import/regression bugs. None of the soft pairs look like the same underlying issue, so all should be rejected.", - "confidence": 0.62, - "canonical_issue_reason": "issue:44521 is the best representative of the largest recurring theme in this set (chat_template / multimodal processor bugs) and has active discussion and references, even though the overall cluster is heterogeneous.", + "summary": "This set is mostly heterogeneous: tokenization, multimodal chat-template, import, runtime, and training issues that only look superficially similar. I found one plausible duplicate family around missing `pad_token_id` on config objects; the rest should stay separate.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43334 is the clearest anchor for the only plausible duplicate family here: an `AttributeError` from missing `pad_token_id` on config objects, with a near-identical sibling in issue:43525.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the most suitable overall issue to anchor this cluster because it is active, well-discussed, and closest to the repeated processor/chat-template failures appearing across nearby items.", + "best_issue_reason": "issue:43334 is the best overall issue to anchor the cluster because it has a concrete, model-loading code-path bug and a plausible duplicate; the other items are mostly unrelated or only broadly similar.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43844", - "right": "issue:44936", - "accept": false, - "reason": "Different bugs: ZeRO-3 gradient growth during training vs trainer.evaluate() failing after train()." - }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Different processor regressions: processor.tokenizer/image handling vs batched video output shape." + "reason": "Different regressions: processor/tokenizer API breakage vs batched video shape handling." }, { "left": "issue:43334", "right": "issue:43525", - "accept": false, - "reason": "Same symptom class (missing pad_token_id) but different configs/models and likely different fixes." + "accept": true, + "reason": "Both report the same underlying `pad_token_id` config attribute bug causing load-time `AttributeError`." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "Unrelated failures: import error for Qwen2_5_VLForConditionalGeneration vs DeepSeek detokenization regression." + "reason": "Unrelated failures: an import error for a missing class vs a detokenization/output issue." }, { - "left": "issue:43388", - "right": "issue:43526", + "left": "issue:43295", + "right": "issue:43873", "accept": false, - "reason": "Both involve labels, but one is gather_for_metrics batching logic and the other is BeitImageProcessorFast reduce_labels." + "reason": "Processor regression and quantization/offloading behavior are different code paths and symptoms." }, { - "left": "issue:43295", - "right": "issue:43873", + "left": "issue:43388", + "right": "issue:43526", "accept": false, - "reason": "Different code paths: custom processor/tokenizer regression vs quantization/offloading behavior." + "reason": "Different bugs in different helpers: metric gathering label truncation vs image processor label reduction." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "Different models and failure modes: zero probabilities in SigLIP2 vs RT-DETR memory not being released." + "reason": "SigLIP2 output anomaly and RT-DETR memory retention are unrelated." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Different issues entirely: FineGrainedFP8Config failure in Qwen3Moe vs a Qwen2.5-VL import error." + "reason": "FP8 config failure and Qwen2.5-VL import failure are different issues." }, { "left": "issue:44514", "right": "issue:44521", "accept": false, - "reason": "Same subsystem, but not the same bug: one crashes on batched padding=False input, the other returns zero assistant masks." + "reason": "Both involve `apply_chat_template`, but one crashes on batched input and the other produces incorrect assistant masks; not the same bug." }, { "left": "issue:38175", "right": "issue:43493", "accept": false, - "reason": "Both touch SigLIP2, but one is zero probabilities and the other is HF-vs-JAX implementation discrepancy." + "reason": "Same model family, but the titles point to different symptoms and not clearly the same fix." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "Different tokenizer/model issues: AutoTokenizer loading failure vs save_pretrained/tokenization failure on Wav2Vec2." + "reason": "Tokenizer load failure vs save/tokenization failure in a different model family." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer spacing after added token is unrelated to embedding reinitialization after resize." + "reason": "Tokenizer spacing behavior and embedding reinitialization are unrelated." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Different problems: dtype drift in evaluate() vs memory inefficiency in Qwen3 MoE training." + "reason": "Training/evaluation dtype change and MoE memory usage are distinct problems." }, { - "left": "issue:43653", - "right": "issue:44112", + "left": "issue:42490", + "right": "issue:43493", "accept": false, - "reason": "Unrelated CI failures: BigBirdTokenizer special-token registration vs GraniteSpeech stale device override test." + "reason": "Local-path import warning and SigLIP2 implementation discrepancy are unrelated." }, { - "left": "issue:42490", - "right": "issue:43493", + "left": "issue:43653", + "right": "issue:44112", "accept": false, - "reason": "Different bugs: local-path import warning vs SigLIP2 implementation discrepancy." + "reason": "Tokenizer special-token registration bug vs a stale CI device-override test." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs/pipeline-related, but they refer to different deprecations and behavior mismatches." + "reason": "Both are docs-related, but they cover different API deprecations and different pages." }, { - "left": "issue:44206", - "right": "issue:45584", + "left": "issue:43450", + "right": "issue:43575", "accept": false, - "reason": "Different failures: unsupported center arg in feature extraction vs empty transcription after align_special_tokens." + "reason": "Batched video shape bug and tensor-parallel OOM are unrelated." }, { - "left": "issue:43450", - "right": "issue:43575", + "left": "issue:44206", + "right": "issue:45584", "accept": false, - "reason": "Different areas: batched video processor shape vs tensor-parallel OOM when loading Qwen2-57B-A14B-Instruct." + "reason": "Feature-extractor center-arg crash and Whisper empty-transcription handling are different failures." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "No shared underlying bug: LoRA merge with extended vocab vs category-count limit runtime error." + "reason": "LoRA/vocabulary merge collapse and category-count overflow are not the same bug." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "Different DAC latent mismatch vs Qwen3 Omni helper-function behavior." + "reason": "DAC STE mismatch and qwen3_omni_moe length helper behavior are different code paths." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "Completely unrelated: DAC latent mismatch vs GraniteSpeech CI device override failure." + "reason": "DAC forward mismatch and a GraniteSpeech CI test issue are unrelated." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Different processor types and failures: audio chat-template sampling-rate default vs batched video shape bug." + "reason": "Audio chat-template sample-rate defaulting and video batching shape are different subsystems." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Different code paths: OmDet-Turbo image sizing vs LwDetrImageLoss AMP/CUDA crash." + "reason": "Processor input-size mismatch and AMP/CUDA loss crash are separate bugs." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Different bugs: embedding dtype change in evaluate() vs pipeline v5 error-message wording." + "reason": "Embedding dtype regression and pipeline docs mismatch are unrelated." }, { - "left": "issue:45290", - "right": "issue:45381", + "left": "issue:37428", + "right": "issue:43425", "accept": false, - "reason": "Different multimodal bugs: tool-call assistant message crash vs wrong vision_position_ids for video input." + "reason": "A specific import error and a Torch version incompatibility are not the same issue." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "Different failures: CPU-offload device mismatch vs Qwen3 Omni helper function semantics." + "reason": "CPU offload device mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." }, { - "left": "issue:37428", - "right": "issue:43425", + "left": "issue:45290", + "right": "issue:45381", "accept": false, - "reason": "Different issues: missing internal import vs Torch 2.10 compatibility." + "reason": "Chat-template tool-call crash and video vision-position-id bug are different multimodal issues." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "Different subsystems: SwitchTransformers sparse-layer construction vs Bamba SDPA/torch.compile failure." + "reason": "Sparse-layer construction bug and Bamba SDPA/compile failure do not share the same root cause." }, { "left": "issue:44448", "right": "issue:44743", "accept": false, - "reason": "Different model regressions: Pegasus v4/v5 output drift vs recurrent state reset in modular_qwen3_5." + "reason": "Output mismatch for Pegasus and recurrent-state reset in modular Qwen3 are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "AMP/CUDA crash and missing auxiliary losses are different model-training bugs." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. The anchor choice around issue:43334 is supported by the matching `pad_token_id` AttributeError in issue:43525, and the summary does not overclaim broader duplication beyond that plausible family. The other pairwise rejections are phrased cautiously and are consistent with the issue titles." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -7611,8 +7528,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -7666,14 +7583,13 @@ "issue:45859" ], "soft_pairs": [ - "issue:44857|issue:45593", "issue:44360|issue:44561", "issue:45200|issue:45588", - "issue:29942|issue:32090", "issue:37428|issue:44351", + "issue:29942|issue:32090", "issue:44206|issue:45823", - "issue:39692|issue:43526", "issue:43611|issue:43931", + "issue:39692|issue:43526", "issue:43335|issue:43653", "issue:43819|issue:44493", "issue:43550|issue:44112", @@ -7695,257 +7611,251 @@ "issue:44977|issue:45229", "issue:42490|issue:43854", "issue:40444|issue:43526", - "issue:43723|issue:44554", "issue:44448|issue:44977", "issue:34689|issue:35141", "issue:43606|issue:44857", + "issue:43723|issue:44554", "issue:44857|issue:45072", "issue:44977|issue:45563" ], - "prepared_review_unit_hash": "b2fcc468b7e72093a408f5bf3d00a3c2b30cca4c7dbdcc736f4df3166184c600", + "prepared_review_unit_hash": "2f25421e8c38744aae2da9fb5ff9b1f023f6ad5de640618830760843e9e5fc3c", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 35, - "serialized_chars": 38195, - "estimated_input_tokens": 9549, - "estimated_eval_tokens": 19354 + "soft_pair_count": 34, + "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": "Mostly a set of loosely related Transformers issues, with many surface-level similarities around v5, flash-attn, loading, and model-specific regressions. Only one clear duplicate pair stands out: the two Qwen3.5 MoE from_pretrained reports (45310 and 45478).", - "confidence": 0.88, - "canonical_issue_reason": "issue:45310 is the best canonical issue: it is the earlier, fuller report and matches issue:45478 on the same Qwen3.5 MoE from_pretrained regression.", + "summary": "Only one soft edge looks like a true duplicate pair; the rest share keywords but describe different bugs, regressions, or subsystems.", + "confidence": 0.97, + "canonical_issue_reason": "issue:45310 is the earlier, fuller report of the same Qwen3.5 MoE `from_pretrained` regression as issue:45478.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the clearest representative of the only true duplicate pair in this set, and its title/body are specific enough to anchor triage.", + "best_issue_reason": "issue:45310 is the clearest duplicate target in this set: specific, reproducible, and matched by the near-identical follow-up report.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44857", - "right": "issue:45593", - "accept": false, - "reason": "Different bugs in different model paths: AMP/CUDA loss crash vs D-FINE auxiliary-loss behavior." - }, { "left": "issue:44360", "right": "issue:44561", "accept": false, - "reason": "Unrelated code paths: DSA ReLU discussion vs trust_remote_code breakage from FX helper removal." + "reason": "Different code paths: DSA indexer/ReLU discussion vs v5 trust_remote_code breakage from `is_torch_fx_available` removal." }, { "left": "issue:45200", "right": "issue:45588", "accept": false, - "reason": "Gemma4 text-only fine-tuning token ids issue is unrelated to flash_attention.py crashing on s_aux=None." + "reason": "Different failures: missing `mm_token_type_ids` for Gemma 4 fine-tuning vs flash-attention crash on `s_aux=None`." }, { - "left": "issue:29942", - "right": "issue:32090", + "left": "issue:37428", + "right": "issue:44351", "accept": false, - "reason": "Flash Attention 2 test failures and Trainer _gpu_broadcast_one NoneType errors are distinct problems." + "reason": "Both are import errors, but for different missing symbols (`_flash_supports_window_size` vs `HybridCache`)." }, { - "left": "issue:37428", - "right": "issue:44351", + "left": "issue:29942", + "right": "issue:32090", "accept": false, - "reason": "Both are import errors, but they are for different missing symbols and different regressions." + "reason": "Flash Attention 2 test failures are unrelated to the Trainer `_gpu_broadcast_one` NoneType broadcast error." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Unsupported feature-extractor arg and device_map/offload mismatch are separate failures." + "reason": "Different subsystems and symptoms: feature extractor `center` arg crash vs Gemma4 device-map mismatch during forward." }, { - "left": "issue:39692", - "right": "issue:43526", + "left": "issue:43611", + "right": "issue:43931", "accept": false, - "reason": "Docs example/model-processor mismatch is unrelated to BeitImageProcessorFast reduce_labels behavior." + "reason": "Model-loading regression from `base_model_prefix` handling is distinct from Qwen3-VL weight-shape mismatch." }, { - "left": "issue:43611", - "right": "issue:43931", + "left": "issue:39692", + "right": "issue:43526", "accept": false, - "reason": "Base-model-prefix loading breakage is a different issue than Qwen3-VL shape mismatches." + "reason": "SigLIP2 docs/example issues are unrelated to BeitImageProcessorFast `reduce_labels` returning one label." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug and BigBirdTokenizer special-token bug are unrelated." + "reason": "SwitchTransformers sparse-layer config bug is unrelated to BigBirdTokenizer special-token decode behavior." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC.from_latents mismatch and unexpected position-id keys are different regressions." + "reason": "DAC `from_latents`/STE mismatch is a different bug from unexpected `position_id` keys across models." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "torch.compile/SDPA failure is unrelated to a stale GraniteSpeech CI test." + "reason": "Different targets: Bamba torch.compile+SDPA failure vs stale GraniteSpeech CI test/device override issue." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "Sparse-layer creation and BEiT legacy-field loading are different code paths." + "reason": "Same broad model family, but different bugs: sparse-layer creation vs Beit legacy-field loading preset failure." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 JIT parsing error is unrelated to qwen3_omni_moe feature-length helper behavior." + "reason": "Python 3.13 JIT parse/indentation import error is unrelated to qwen3_omni_moe feature-length helper behavior." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Same title, same Qwen3.5 MoE from_pretrained regression, and clearly the same underlying loading bug." + "reason": "Near-identical reports of the same Qwen3.5 MoE `from_pretrained` regression on transformers>=5.4.0." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "A SwitchTransformers config bug is unrelated to the qwen3_omni_moe helper function issue." + "reason": "SwitchTransformers config bug and qwen3_omni_moe feature-length helper bug are unrelated." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 error messaging and Trainer broadcast NoneType are unrelated." + "reason": "LayoutLMV3 box/clarity error is unrelated to Trainer `_gpu_broadcast_one` NoneType broadcast failure." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Both mention Qwen3.5-related paths, but the crashes occur in different functions and failure modes." + "reason": "Different flash-attention edge cases: sink-less `s_aux=None` crash vs CPU-offload incompatibility in Qwen3_5MoE rotary embedding." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "A documentation/example error is not the same bug as a widespread unexpected-key regression." + "reason": "SigLIP2 documentation/model mismatch is unrelated to unexpected `position_id` keys." }, { "left": "issue:44292", "right": "issue:45520", "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime failure and flash_attn import KeyError on Python 3.13 are different issues." + "reason": "Qwen-3-8B-NVFP4 runtime failure is unrelated to the Python 3.13 `flash_attn` import key error." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 zero-probability output is unrelated to unexpected key warnings with position ids." + "reason": "Zero-probability outputs in SigLIP2 are a different symptom from unexpected `position_id` key warnings." }, { "left": "issue:43335", "right": "issue:45072", "accept": false, - "reason": "Different model families and different failure modes; not the same bug." + "reason": "Different failure modes: SwitchTransformers sparse-layer config vs dtype mismatches in bfloat16 inference." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Pegasus v4/v5 output drift and Deepseek tokenizer regression are different model-specific problems." + "reason": "Different regressions in different models: Pegasus v4/v5 output mismatch vs Deepseek tokenizer correctness change." }, { "left": "issue:44462", "right": "issue:44561", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to FX-helper removal breaking trust_remote_code models." + "reason": "AutoTokenizer ignoring `tokenizer.json` is unrelated to trust_remote_code breakage from removed `is_torch_fx_available`." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Different regressions: trust_remote_code import breakage versus tokenizer correctness in Deepseek." + "reason": "Different areas: remote-code utility removal vs Deepseek tokenizer output regression." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "SwitchTransformers sparse-layer config and D-FINE auxiliary-loss behavior are unrelated." + "reason": "SwitchTransformers sparse-layer config bug is unrelated to D-FINE auxiliary-loss handling." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "BitNet accelerate-loading unpacking bug is unrelated to a stale device-override CI test." + "reason": "BitNet accelerate-loading unpack bug is unrelated to GraniteSpeech stale CI device-override test." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Qwen3.5 flash-attention generation issue and Gemma4 multi-GPU OOM are different problems." + "reason": "Qwen3.5 flash-attention generation bug is unrelated to Gemma4 multi-GPU CUDA OOM." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Trailing-slash import warning and GLM-4.7-Flash unit-test load failure are not the same bug." + "reason": "Trailing-slash local model import warning is unrelated to GLM-4.7-Flash loading failure in tests." }, { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "IterableDataset multi-image finetuning failure is unrelated to BeitImageProcessorFast label reduction." - }, - { - "left": "issue:43723", - "right": "issue:44554", - "accept": false, - "reason": "AutoTokenizer v5 loading breakage and MPS attention correctness are distinct issues." + "reason": "IterableDataset multi-image finetuning failure is unrelated to BeitImageProcessorFast `reduce_labels` bug." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different models and different regressions; not mergeable as one fix." + "reason": "Different model/output regressions: Pegasus v4/v5 mismatch vs Qwen3.5 flash-attention generation issue." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Llama 3.2 Vision loading regression and output-embedding reinit on resize are separate bugs." + "reason": "Llama 3.2 Vision loading breakage and token-embedding reinit on `post_init` are different bugs." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "CPU offload device mismatch and float16 AMP loss crash are unrelated." + "reason": "CPU offload device mismatch in bark-small is unrelated to LwDetrImageLoss AMP/CUDA crash." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "Tokenizer loading in v5 and MPS attention correctness are different problems." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "These are different dtype/training-inference issues affecting different models and paths." + "reason": "Different failures: loss/AMP crash vs dtype mismatch in inference." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "A flash-attention generation bug is not the same as a stale warning about num_return_sequences." + "reason": "Qwen3.5 flash-attention generation bug is unrelated to the stale `num_return_sequences` warning in paged generate()." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analysis is grounded in the packet. The only accepted duplicate pair, issue:45310 and issue:45478, has identical titles and clearly matches the same Qwen3.5 MoE from_pretrained regression. The remaining soft-edge rejections are conservative and appropriately avoid overstating similarity." + "feedback": "The canonical duplicate choice is well grounded: issue:45310 and issue:45478 have effectively the same title and clearly refer to the same Qwen3.5 MoE from_pretrained regression. The remaining soft-edge rejections are conservative and supported by the packet-level titles/symptoms, without overstating evidence." }, "evaluator_used": true, "retried": false, @@ -7961,8 +7871,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -8024,10 +7934,10 @@ "issue:45372|issue:45588", "issue:37428|issue:44987", "issue:29942|issue:35141", - "issue:45459|issue:45712", "issue:43873|issue:44493", - "issue:45245|issue:45563", + "issue:45459|issue:45712", "issue:32090|issue:33357", + "issue:45245|issue:45563", "issue:43412|issue:45588", "issue:43873|issue:45229", "issue:41720|issue:45229", @@ -8041,219 +7951,219 @@ "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:44448|issue:44466", + "issue:43278|issue:43381", + "issue:45588|issue:45823" ], - "prepared_review_unit_hash": "9e821cf196fe3a722940f2378bd666fa44a935e2d717a32cfe162189146285c9", + "prepared_review_unit_hash": "8e9fcfb3eaf43d6f226da521c48ee5d0c3fc8045d73de8800f7e9f5a569d8b78", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36375, - "estimated_input_tokens": 9094, - "estimated_eval_tokens": 18444 + "serialized_chars": 36365, + "estimated_input_tokens": 9092, + "estimated_eval_tokens": 18440 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is highly heterogeneous: most pairs are only superficially similar and do not look like the same underlying bug or change. I would not merge any of the soft-edge candidates.", - "confidence": 0.19, - "canonical_issue_reason": "If a single issue must anchor this set, issue:45698 is the cleanest well-scoped regression report, but overall this cluster does not have a strong canonical duplicate target because the items span unrelated subsystems and failure modes.", + "summary": "This cluster is broadly heterogeneous. A few items share themes like tokenizer/save-load regressions or Gemma4 processor-loading failures, but none of the soft pairs look like the same concrete bug/change, so I reject all duplicate candidates.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45372 is the most central representative: it is a current Gemma4 processor-loading regression, has downstream references, and sits closest to the small number of related Gemma4/processor items in the set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45698 is the best representative issue because it is concrete, reproducible, and framed as a specific save/load regression, which makes it a better triage anchor than the more vague or highly specialized reports.", + "best_issue_reason": "issue:45372 is the best single anchor for this set because it is specific, recent, open, and most likely to attract related follow-up reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44568", "right": "issue:45491", "accept": false, - "reason": "Different bugs: tokenizer special-token insertion vs Gemma3 NaN embeddings with mixed-length batching." + "reason": "Tokenizer special-token behavior vs Gemma3 NaN embeddings are unrelated bugs." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Both involve save/load, but one is config token IDs and the other is wrong custom-module resolution; different code paths and symptoms." + "reason": "Both involve save/load flows, but one is config token IDs and the other is wrong custom module resolution." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer spacing bug vs Trainer distributed broadcast TypeError; unrelated failures." + "reason": "Tokenizer spacing after added tokens is unrelated to Trainer broadcast NoneType errors." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint key mismatch is unrelated to qwen3_omni_moe feature helper behavior." + "reason": "Checkpoint key mismatch and multimodal feature-length behavior are different code paths." }, { "left": "issue:43550", "right": "issue:45083", "accept": false, - "reason": "torch.compile/SDPA failure is not the same as an output-length helper bug." + "reason": "torch.compile/SDPA failure is unrelated to the feature-length helper issue." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "save_pretrained/tokenization failure is not the same as a bare-except swallowing errors in Fuyu conversion." + "reason": "Wav2Vec2 save/tokenization failure and Fuyu bare-except masking are distinct tokenizer issues." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Gemma4 processor import dependency issue vs flash_attention crash on s_aux=None; different concrete problems." + "reason": "Both touch Gemma4-related loading/runtime, but one is an import dependency break and the other is a flash-attention attribute crash." }, { "left": "issue:37428", "right": "issue:44987", "accept": false, - "reason": "ImportError for a flash-attention helper vs a model loading failure for a specific repo; not the same defect." + "reason": "Missing flash-attention utility import is not the same as failing to load a specific model repository." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Failing FA2 tests and tied-weight reinitialization are unrelated regressions." - }, - { - "left": "issue:45459", - "right": "issue:45712", - "accept": false, - "reason": "One is exception masking in tokenizer error handling; the other is stale dummy PT classes in repo inspection." + "reason": "Flash Attention 2 test failures are unrelated to embedding reinitialization on resize." }, { "left": "issue:43873", "right": "issue:44493", "accept": false, - "reason": "Quantization/offloading behavior vs unexpected position-id keys; different symptom and subsystem." + "reason": "Quantized offloading behavior and unexpected position-id keys are different defects." }, { - "left": "issue:45245", - "right": "issue:45563", + "left": "issue:45459", + "right": "issue:45712", "accept": false, - "reason": "Category-cardinality runtime limit is unrelated to a stale generate() warning." + "reason": "Masked tokenizer decode errors and leftover dummy classes are unrelated maintenance issues." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "GPU broadcast NoneType error vs MacOS bus error with a CLIP model; not the same underlying bug." + "reason": "Trainer NoneType broadcast error is unrelated to a MacOS bus error on a CLIP model." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-cardinality runtime error is unrelated to a stale warning in paged generate()." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Executorch export/erfinv/training-guard issue is distinct from a flash-attention s_aux=None crash." + "reason": "Executorch export failures and flash-attention sink-less crash are different concrete problems." }, { "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Offloading-with-quantization and multi-GPU OOM are both memory-related but not the same code-path problem." + "reason": "Quantization/offloading behavior is not the same as multi-GPU OOM on Gemma4." }, { "left": "issue:41720", "right": "issue:45229", "accept": false, - "reason": "Auto device mapping cudaErrorAssert is a different failure mode from plain OOM." + "reason": "A800 cudaErrorAssert with auto device mapping is not the same as a multi-GPU OOM." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "Embedding dtype drift in eval is unrelated to undefined video-branch variables in multimodal token counting." + "reason": "Embedding dtype drift and undefined video-branch helpers are unrelated model issues." }, { "left": "issue:43976", "right": "issue:45440", "accept": false, - "reason": "Python version compatibility issue vs divergence from remote DeepSeekV3 implementation; unrelated." + "reason": "Python version compatibility and DeepSeek implementation divergence are different failures." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "Different models, different frameworks, different errors." + "reason": "German DistilBertModel issue and torch.jit GeneratorExp export failure are unrelated." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Both touch Gemma4 loading, but one is an import dependency problem and the other is a missing _tokenizer attribute in serve; not the same bug." + "reason": "Both involve Gemma4Processor, but the reported failures point to different root causes and code paths." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Dummy Conv2d placeholder in Sam3PixelDecoder vs nonsensical Siglip2 outputs; no shared concrete defect." + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to SigLIP2 producing nonsensical outputs." }, { "left": "issue:42490", "right": "issue:43575", "accept": false, - "reason": "Trailing-slash import warning is unrelated to Qwen2 tensor-parallel OOM." + "reason": "Trailing-slash import warning and tensor-parallel OOM are unrelated." }, { "left": "issue:28218", "right": "issue:29127", "accept": false, - "reason": "Tokenizer added-space regression vs layoutlmv3 box error messaging; unrelated." + "reason": "Tokenizer added-token spacing and LayoutLMv3 box-validation messaging are different issues." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "Dummy layer placeholder vs RT-DETR memory not being released are different bugs." + "reason": "Sam3PixelDecoder dummy layer issue is unrelated to RT-DETR memory not releasing." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma4 dependency import failure is not the same as protobuf exception masking in tokenizer decode paths." + "reason": "Gemma4 processor import failure and protobuf-error masking in tokenizers are not the same bug." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "AMP/CUDA loss crash is unrelated to flash-attention sink-less-model crash." - }, - { - "left": "issue:44448", - "right": "issue:44466", - "accept": false, - "reason": "Different-output regression across versions vs inconsistent lm_head serialization; same broad area but not the same underlying bug." + "reason": "AMP/CUDA loss crash and flash-attention sinkless-model crash are different runtime failures." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "Tokenizer special-token registration issue vs recurrent-state reset in cache handling; different code paths." + "reason": "BigBird special-token registration and Qwen3 recurrent-state reset are unrelated." }, { - "left": "issue:45588", - "right": "issue:45823", + "left": "issue:44448", + "right": "issue:44466", "accept": false, - "reason": "Flash-attention attribute crash and Gemma4 PLE device mismatch are separate forward-path failures." + "reason": "Version output differences and lm_head serialization inconsistency are distinct model-state bugs." }, { "left": "issue:43278", "right": "issue:43381", "accept": false, - "reason": "dtype change in evaluation is unrelated to gradient checkpointing being disallowed in eval mode." + "reason": "Training/eval dtype change and gradient-checkpointing-in-eval are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Flash-attention s_aux=None crash and Gemma4 PLE device mismatch are separate Gemma4 runtime bugs." } ] }, @@ -8272,8 +8182,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -8357,208 +8267,208 @@ "issue:29942|issue:36010", "issue:29127|issue:33357" ], - "prepared_review_unit_hash": "513de4dca337028a817276c65529fccefdfb4d4fb40a2843af92eabeffa4a45b", + "prepared_review_unit_hash": "ee0a4c73a5d587a76ac5f59388e886383b5527958b8abbd48ef3f349c652a9d6", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36301, - "estimated_input_tokens": 9076, - "estimated_eval_tokens": 18408 + "serialized_chars": 36306, + "estimated_input_tokens": 9077, + "estimated_eval_tokens": 18410 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No duplicate cluster emerges here: the items span unrelated tokenizer, vision, trainer, quantization, documentation, and export issues. All soft pairs are false positives and should stay separate.", - "confidence": 0.98, - "canonical_issue_reason": "No single canonical issue fits this set; the titles and bug reports cover distinct code paths and model families, and the soft-linked pairs do not describe the same underlying defect.", + "summary": "This cluster is a broad set of mostly unrelated Transformers issues that only look superficially similar by text embedding. None of the proposed soft edges are true duplicates, so all should be rejected. There is no meaningful canonical issue for the cluster.", + "confidence": 0.92, + "canonical_issue_reason": "No canonical issue: the items span different subsystems and failure modes (tokenization, quantization, training, vision, docs, export, offload), so there is no single underlying bug to represent the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "There is no strong global best issue representative for the cluster because the items are heterogeneous and none subsumes the others as a duplicate target.", + "best_issue_reason": "No global best issue: the soft matches are all broad similarity false positives rather than one coherent bug family, so no issue is a strong dedupe anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "Different regressions in different subsystems: image feature extraction vs video-input handling for Qwen models." + "reason": "Different regressions in different model paths: LasrFeatureExtractor center-arg handling vs Qwen video-input processing." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Both concern tokenizers, but one is a decoder-type mismatch in v5 while the other is AutoTokenizer ignoring tokenizer.json; different bugs." + "reason": "Both involve tokenizers, but one is a decoder-type mismatch in v5 while the other is AutoTokenizer ignoring tokenizer.json; not the same bug." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 inference mismatch and RT-DETR memory reclamation are unrelated problems." + "reason": "Unrelated failures: SigLIP2 inference corruption vs RT-DETR model memory not being released." }, { "left": "issue:25251", "right": "issue:32090", "accept": false, - "reason": "Pipeline output nesting and Trainer GPU broadcast TypeError are unrelated code paths." + "reason": "Top-k pipeline output nesting and trainer _gpu_broadcast_one NoneType errors are unrelated code paths." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2 NER/padding crashes and Wav2Vec2 save_pretrained/tokenization failures are different tokenizer issues." + "reason": "Both are tokenizer-related, but the concrete failures differ: LayoutLMv2 NER/padding crash vs Wav2Vec2 save_pretrained/tokenization failure." }, { "left": "issue:44451", "right": "issue:45520", "accept": false, - "reason": "Model load failure for ScandiBERT and a Python 3.13 flash_attn import KeyError are unrelated." + "reason": "Different root causes: model loading failure for ScandiBERT vs Python 3.13 flash_attn import-key issue." }, { "left": "issue:42915", "right": "issue:42994", "accept": false, - "reason": "Both mention quantization, but one is an FP8 loading failure and the other is a quantized-model saving bug." + "reason": "Both touch quantization, but one is Qwen3MoE with FineGrainedFP8Config and the other is quantized model saving; not the same defect." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "MLukeTokenizer task errors do not match Wav2Vec2 save/tokenization failures." + "reason": "Different tokenizer/model failures affecting different tasks and APIs; no shared underlying bug is evident." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "Gemma eos_token_id inconsistency and Qwen3.5 flash-attention generation issues are different defects." + "reason": "Eos-token config mismatch is not the same issue as Qwen3.5 flash-attention generation failure." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export with torch_compilable_check is unrelated to Wav2Vec2 tokenization/save_pretrained behavior." + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 tokenization/save_pretrained problems." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Both are model-implementation mismatches, but for different architectures and different misbehaviors." + "reason": "Both concern model implementation divergence, but they affect different architectures and different mechanisms; not mergeable as one fix." }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "Runtime special-token post-processing update bug and Gemma3 NaN embeddings are distinct issues." + "reason": "Runtime special-token post-processing is unrelated to Gemma3 NaN embeddings from sliding-window attention." }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Flash Attention 2 test failures and void segmentation-map processing are unrelated." + "reason": "Flash Attention 2 test failures and void segmentation map image-processing failures are separate bugs." }, { "left": "issue:43257", "right": "issue:43278", "accept": false, - "reason": "Qwen3 MoE weight conversion/loading and embedding dtype drift in eval are different bugs." + "reason": "Different issues: MoE weight conversion with accelerate+deepspeed vs embedding dtype changing between train/eval." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "MacOS CLIP bus error and SigLIP2 documentation example errors are not the same underlying issue." + "reason": "Both mention CLIP/SigLIP-family examples, but one is a Mac bus error on a community model and the other is a documentation example mismatch/quantization issue." }, { "left": "issue:44610", "right": "issue:45412", "accept": false, - "reason": "Processor input-size mismatch and RT-DETR memory leak are unrelated." + "reason": "Processor output size mismatch and RT-DETR memory leak are unrelated." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "LayoutLMv3 error-message clarity and TrainerState token counter not updating are different concerns." + "reason": "LayoutLMv3 error-message clarity and TrainerState token-count tracking are different issues." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Trailing-slash import warning and quantization offloading behavior are unrelated." + "reason": "Local-path import warning and quantization offloading behavior are not the same bug." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "NoneType broadcast error in Trainer and DeepSpeed OOM with Adafactor are separate failures." + "reason": "Trainer NoneType broadcast error and DeepSpeed Adafactor OOM are different failure modes in different layers." }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "auto_docstring annotation handling and Gemma 4 processor import failure are unrelated." + "reason": "auto_docstring annotation handling and Gemma 4 processor loading/import error are unrelated." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "DeepseekV3 implementation divergence and D-FINE auxiliary-loss handling are different model-specific bugs." + "reason": "DeepseekV3MoE implementation drift is not the same as D-FINE auxiliary-loss logic when denoising is off." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "Qwen2.5-VL import failure and a generic dependency issue are not the same defect." + "reason": "Qwen2.5-VL import failure is not the same as a generic dependency issue in transformers." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "BigBirdTokenizer special-token registration and AutoTokenizer failing on FastSpeech2ConformerTokenizer are different tokenizer problems." + "reason": "BigBird mask-token registration/empty decode and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are different tokenizer bugs." }, { "left": "issue:43761", "right": "issue:45859", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are unrelated." + "reason": "CLIPVisionModel hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are unrelated model-forward issues." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug and Kosmos2.5 long-OCR index error are different processing paths." + "reason": "BeitImageProcessorFast reduce_labels behavior and Kosmos2.5 OCR index errors are different image-processing problems." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Eval-mode gradient checkpointing and resume_from_checkpoint key mismatch are separate trainer behaviors." + "reason": "Gradient checkpointing in eval mode and resume_from_checkpoint key mismatch are not the same defect." }, { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Image processor label reduction and LayoutLMv2 tokenizer crashes are unrelated." + "reason": "ImageProcessorFast label reduction and LayoutLMv2 tokenizer crashes are different components and failure modes." }, { "left": "issue:29942", "right": "issue:36010", "accept": false, - "reason": "Flash Attention test failures and GenerationMixin import error are not the same issue." + "reason": "Flash Attention 2 test failures are unrelated to an ImportError for GenerationMixin." }, { "left": "issue:29127", "right": "issue:33357", "accept": false, - "reason": "LayoutLMv3 error messaging and CLIP bus error on MacOS are unrelated." + "reason": "LayoutLMv3 error-message clarity and a MacOS CLIP bus error are unrelated." } ] }, @@ -8577,8 +8487,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -8634,11 +8544,11 @@ "soft_pairs": [ "issue:43334|issue:43531", "issue:44977|issue:45245", - "issue:44464|issue:45491", "issue:29127|issue:35141", + "issue:44464|issue:45491", "issue:42915|issue:43931", - "issue:43262|issue:44206", "issue:44442|issue:45072", + "issue:43262|issue:44206", "issue:43526|issue:44857", "issue:42490|issue:43381", "issue:44062|issue:45907", @@ -8660,190 +8570,196 @@ "issue:42490|issue:44743", "issue:33290|issue:34567" ], - "prepared_review_unit_hash": "18367ab8a38527be21667d6620858b23ff03cfd066553c0824b24b06a4a52efd", + "prepared_review_unit_hash": "4dfbf2e175a9b36c154645749ed5b3373096da7a26b298968905913070f7320e", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35877, - "estimated_input_tokens": 8970, - "estimated_eval_tokens": 18196 + "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": "No duplicate matches; the soft pairs span distinct bugs across loading, generation, processors, config APIs, and model-specific failures.", - "confidence": 0.88, - "canonical_issue_reason": "Issue 45698 is the clearest standalone bug report in the set: it has active discussion and inbound references, and it describes a concrete save/load regression with reproducible behavior.", + "summary": "The items are mostly unrelated one-off bugs/regressions across different subsystems; the soft pairs share superficial vocabulary but not the same concrete failure mode. I would reject all soft edges.", + "confidence": 0.96, + "canonical_issue_reason": "issue:45698 is the broadest, most general loading/saving bug in the set and is still open; the rest are mostly model-specific or subsystem-specific regressions that do not form a clear duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44387 is the strongest issue report overall, with the richest discussion and a well-scoped, high-impact memory regression description.", + "best_issue_reason": "issue:45698 is the best representative if one issue must anchor this cluster: it concerns core from_pretrained/save_pretrained behavior and has broader reuse potential than the other narrowly scoped bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43334", "right": "issue:43531", "accept": false, - "reason": "Different bugs: missing pad_token_id on Qwen3-VL config vs sliding_window behavior in Qwen3-MoE." + "reason": "Different Qwen issues: missing `pad_token_id` on Qwen3-VL vs `sliding_window` behavior on Qwen3-MoE; not the same bug." }, { "left": "issue:44977", "right": "issue:45245", "accept": false, - "reason": "Unrelated failures: flash-attention generation problem vs category-count limit error in a different model path." + "reason": "Flash-attention generation regression vs a generic category-count runtime limit; unrelated failure modes." }, { - "left": "issue:44464", - "right": "issue:45491", + "left": "issue:29127", + "right": "issue:35141", "accept": false, - "reason": "Compiled-forward generation inconsistency is not the same as Gemma3 sliding-window NaN embeddings." + "reason": "LayoutLMv3 box-clarity error message vs token-embedding reinitialization after resize; different code paths and symptoms." }, { - "left": "issue:29127", - "right": "issue:35141", + "left": "issue:44464", + "right": "issue:45491", "accept": false, - "reason": "LayoutLMv3 error-message clarity and embedding reinitialization after resize are unrelated issues." + "reason": "Compiled chunked generation inconsistency vs Gemma3 NaN embeddings from sliding-window/all-padding windows; not the same defect." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "FP8 config failure for Qwen3Moe is distinct from Qwen3-VL weight-shape mismatch on load." + "reason": "Qwen3Moe FineGrainedFP8Config failure vs Qwen3-VL weight-shape mismatch; separate model-loading problems." }, { - "left": "issue:43262", - "right": "issue:44206", + "left": "issue:44442", + "right": "issue:45072", "accept": false, - "reason": "Audio chat-template sampling-rate default and unsupported center arg in a feature extractor are different processor bugs." + "reason": "Tokenizer loading failure for FastSpeech2ConformerTokenizer vs dtype mismatch in SwitchTransformers/TimmWrapperModel inference." }, { - "left": "issue:44442", - "right": "issue:45072", + "left": "issue:43262", + "right": "issue:44206", "accept": false, - "reason": "Tokenizer loading failure and dtype mismatch inference are separate code paths with different root causes." + "reason": "Audio chat-template sampling-rate default issue vs unsupported `center` argument in a different feature extractor." }, { "left": "issue:43526", "right": "issue:44857", "accept": false, - "reason": "BeitImageProcessorFast label reduction and LwDetrImageLoss AMP crash are not the same defect." + "reason": "BEiTImageProcessorFast label reduction bug vs LwDetrImageLoss AMP/CUDA crash; unrelated subsystems." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Trailing-slash local model warning is unrelated to gradient-checkpointing restrictions in eval mode." + "reason": "Trailing-slash local model import warning vs gradient checkpointing being disallowed in eval mode; not the same bug." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "AddedToken keyword collision and list_repo_files revision forwarding are different API issues." + "reason": "`AddedToken` keyword collision vs `list_repo_files` not forwarding `revision`; different APIs and failures." }, { "left": "issue:43232", "right": "issue:43388", "accept": false, - "reason": "Generation kwargs sync handling and gather_for_metrics label truncation affect different subsystems." + "reason": "Generation kwargs update after `sync_gpus` vs metric gathering dropping tuple labels; separate bugs." }, { "left": "issue:43441", "right": "issue:45588", "accept": false, - "reason": "Both mention flash attention, but they target different model behaviors and different concrete crashes." + "reason": "Both involve flash attention, but one is a Ministral-3 model regression and the other is an `s_aux=None` crash in a different integration path." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Qwen3_5 recurrent-state reset and Whisper batch_decode skip_special_tokens are unrelated." + "reason": "Recurrent state reset in Qwen3.5 MoE vs Whisper `batch_decode` ignoring `skip_special_tokens`; no shared underlying bug." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "CLIPVisionModel hidden_states handling is a different bug from ModelOutput key assignment with prior None values." + "reason": "CLIPVisionModel not returning hidden states vs ModelOutput key assignment when a key was previously `None`; different abstractions." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Custom model initialization breakage in v5 is not the same as patchtsmixer post_init being disallowed." + "reason": "Transformers 5.0 custom model init breakage vs `patchtsmixer` `post_init` contract issue; not the same problem." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "Tensor-parallel OOM and resume_from_checkpoint key mismatch are unrelated failures." + "reason": "TP loading OOM for a Qwen2 model vs checkpoint resume key mismatch; unrelated." }, { "left": "issue:45003", "right": "issue:45440", "accept": false, - "reason": "Unsafe sys.modules access and DeepseekV3MoE divergence concern different code paths and symptoms." + "reason": "Unsafe `sys.modules` access in `modeling_utils` vs DeepSeekV3MoE implementation divergence; different code paths." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Both involve OOM, but one is TP loading and the other is int4 quantization memory regression; not the same bug." + "reason": "Both mention OOM, but one is TP model loading and the other is int4 quantization reserved-memory growth; not mergeable as one bug." }, { "left": "issue:43723", "right": "issue:45479", "accept": false, - "reason": "Tokenizer-loading regression and zero-loss classification bug are unrelated." + "reason": "AutoTokenizer loading issue in v5 vs degenerate zero-loss sequence classification; unrelated." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "AutoTokenizer loading failure and bare-except error handling in Fuyu conversion are different issues." + "reason": "Tokenizer loading failure vs bare `except` in Fuyu tensor conversion; no shared underlying defect." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "Hidden_states=None bug and torch.compile attention-kernel divergence are unrelated." + "reason": "Hidden-states regression in CLIPVisionModel vs attention-kernel divergence under `torch.compile`; distinct issues." }, { "left": "issue:44479", "right": "issue:45290", "accept": false, - "reason": "Video-input regression in VL models and assistant tool-call chat-template crash are different bugs." + "reason": "Video-input regression for Qwen-family models vs chat-template tokenization crash on tool-call assistant messages; different layers." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Both are config/API mismatches, but for different model families and different missing fields." + "reason": "Gemma3TextConfig API discrepancy vs Llama4Config missing `pad_token_id`; similar theme, but not the same bug." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Saving dequantized models and loading the wrong custom module after save_pretrained are distinct serialization problems." + "reason": "Both touch model save/load, but one is dequantized Ministral save failure and the other is wrong custom module loading after save/load; too different to merge." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "FastSpeech2ConformerTokenizer loading and Perceiver non-default-resolution failures are unrelated." + "reason": "Tokenizer loading failure vs Perceiver non-default-resolution image classification failure; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local path import warning vs Qwen3.5 recurrent-state reset bug; no common failure mode." }, { "left": "issue:33290", "right": "issue:34567", "accept": false, - "reason": "Deepspeed Adafactor OOM and TrainerState token counter not updating are separate bugs." + "reason": "DeepSpeed Adafactor OOM vs `num_input_tokens_seen` not updating; entirely separate training issues." } ] }, @@ -8862,8 +8778,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -8919,11 +8835,11 @@ "soft_pairs": [ "issue:34689|issue:43611", "issue:39692|issue:45491", - "issue:43931|issue:45216", "issue:33290|issue:35141", + "issue:43931|issue:45216", "issue:33290|issue:34689", - "issue:43493|issue:43827", "issue:43381|issue:43873", + "issue:43493|issue:43827", "issue:44077|issue:44479", "issue:43316|issue:45759", "issue:43720|issue:44442", @@ -8937,29 +8853,29 @@ "issue:44855|issue:45137", "issue:43316|issue:44877", "issue:43540|issue:44514", - "issue:43673|issue:44743", "issue:42175|issue:43316", - "issue:43404|issue:43441", + "issue:43673|issue:44743", "issue:43065|issue:44617", + "issue:43404|issue:43441", "issue:44162|issue:44792", "issue:43295|issue:43825", "issue:25251|issue:28282", "issue:43388|issue:43688" ], - "prepared_review_unit_hash": "8ecb60070cb6f3f2283bfb9120cf755b079ca7477bdb02b847ce0c7b315d3c21", + "prepared_review_unit_hash": "1a7b2f9b063d6aaf2222e6d3d8f35b2cb02f0997eaa0d4afcecb595a62dce6fa", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36333, + "serialized_chars": 36336, "estimated_input_tokens": 9084, "estimated_eval_tokens": 18424 }, @@ -8968,180 +8884,180 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items form a heterogeneous set of Transformers bugs, regressions, and documentation issues rather than one duplicate cluster. None of the proposed soft pairs are the same underlying defect, so no merges are justified.", - "confidence": 0.19, - "canonical_issue_reason": "If a representative issue must be picked, issue:34689 is a clear, concrete loading regression with a specific model/version impact; however, the overall set is too diverse to support a true canonical duplicate.", + "summary": "This cluster is mostly a false-positive semantic neighborhood: the issues span unrelated regressions, model families, and API surfaces. I found no pair that looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No true duplicate center emerges; if one issue must be chosen, issue 34689 is a clear, well-described regression report and a reasonable representative of the cluster\u2019s broad 'model loading breakage' theme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:34689 is one of the strongest standalone reports here: specific repro, user-visible breakage, and a well-scoped model-loading regression.", + "best_issue_reason": "Issue 34689 is the most concrete and broadly representative bug report in an otherwise heterogeneous set, but it is not a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:34689", "right": "issue:43611", "accept": false, - "reason": "Both are loading regressions, but one is Llama 3.2 Vision-specific while the other is a generic base_model_prefix loading issue in v5; different code paths and fixes." + "reason": "Both involve loading regressions, but one is a specific Llama 3.2 Vision breakage in 4.46.2 and the other is a v5 base_model_prefix loading change; different code paths and failure modes." }, { "left": "issue:39692", "right": "issue:45491", "accept": false, - "reason": "Documentation/example errors for SigLIP2 versus a Gemma3 GPU NaN embedding bug; unrelated symptoms and subsystems." + "reason": "Different models and different bugs: SigLIP2 docs/example errors versus Gemma3 NaN embeddings during mixed-length batching." }, { - "left": "issue:43931", - "right": "issue:45216", + "left": "issue:33290", + "right": "issue:35141", "accept": false, - "reason": "Both mention Qwen3.5/3-VL model handling, but one is a shape-mismatch load failure and the other is a save_pretrained checkpoint regression; not the same bug." + "reason": "OOM with Adafactor + DeepSpeed is unrelated to token embedding resize reinitialization in post_init." }, { - "left": "issue:33290", - "right": "issue:35141", + "left": "issue:43931", + "right": "issue:45216", "accept": false, - "reason": "Adafactor+DeepSpeed OOM and embedding reinitialization in post_init are different failure modes with no shared code-path." + "reason": "One is a weight-shape mismatch during loading, the other is an incorrect save_pretrained checkpoint for Qwen3.5; not the same underlying bug." }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "Optimizer/DeepSpeed memory issue versus model loading regression; only broad ML-training context overlaps." + "reason": "Different subsystems and symptoms: optimizer/deepspeed memory blowup versus model-loading regression for a specific vision model." }, { - "left": "issue:43493", - "right": "issue:43827", + "left": "issue:43381", + "right": "issue:43873", "accept": false, - "reason": "SigLIP2 implementation mismatch is unrelated to docs still referencing pipeline() after v5 removals." + "reason": "Gradient checkpointing in eval mode is unrelated to quantization/offloading behavior." }, { - "left": "issue:43381", - "right": "issue:43873", + "left": "issue:43493", + "right": "issue:43827", "accept": false, - "reason": "Gradient checkpointing in eval mode and quantized offloading behavior are distinct training/runtime issues." + "reason": "SigLIP2 implementation discrepancy and stale pipeline documentation are different issues." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "PatchTSMixer post_init policy and Qwen video-input regression are unrelated model-specific bugs." + "reason": "PatchTSMixer post_init validation and Qwen video-input regression are unrelated." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Both involve config APIs, but one is a Gemma3TextConfig discrepancy and the other is unwrapping text_config for Qwen multimodal configs; different semantics and fixes." + "reason": "Both touch config APIs, but one is a Gemma3TextConfig inconsistency and the other is AutoModelForCausalLM unwrapping composite multimodal text_config; different fixes." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight loading failure and FastSpeech2ConformerTokenizer auto-loading failure are unrelated." + "reason": "BitNet accelerate loading with packed weights is unrelated to AutoTokenizer failing for a custom tokenizer." }, { "left": "issue:41628", "right": "issue:45372", "accept": false, - "reason": "Both are import errors, but they stem from different missing symbols in different processor/model stacks." + "reason": "Both are import errors, but they concern different missing symbols and different dependency chains." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "trust_remote_code returning missing config fields is not the same as GPTNeoX rotary_pct not persisting on reload." + "reason": "Trust-remote-code config field loss is not the same as GPTNeoX rotary_pct resetting on reload." }, { "left": "issue:43749", "right": "issue:43931", "accept": false, - "reason": "FSDP CPU RAM efficient loading is a distributed-loading regression; Qwen3-VL weight mismatch is a model checkpoint shape issue." + "reason": "FSDP CPU RAM efficient loading failure is not the same as a specific model weight-shape mismatch." }, { "left": "issue:34689", "right": "issue:36010", "accept": false, - "reason": "Llama model loading regression versus a GenerationMixin import error; different APIs and root causes." + "reason": "Different loading errors with different root causes; one is a version regression for Llama 3.2 Vision, the other is a missing GenerationMixin import path." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantized offloading behavior and Gemma-3 eos_token_id inconsistency are unrelated." + "reason": "Quantization/offloading behavior and Gemma-3 eos_token_id inconsistency are unrelated." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in an example script is unrelated to an unrecognized model/config.json failure." + "reason": "Dataset loading in a training script is unrelated to unrecognized model_type in config.json." }, { "left": "issue:33290", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed Adafactor OOM and CustomTrainer compute_loss signature mismatch are different training issues." + "reason": "OOM in DeepSpeed/Adafactor is unrelated to CustomTrainer.compute_loss signature mismatch." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 import/parsing problem in DebertaV2Model is unrelated to a DeepSpeed ZeRO3 deque IndexError." + "reason": "Python 3.13 IndentationError on import is unrelated to a DeepSpeed ZeRO3 deque underflow." }, { "left": "issue:43316", "right": "issue:44877", "accept": false, - "reason": "API discrepancy in Gemma3TextConfig versus strict config blocking granite_speech loading; related to configs broadly but not the same defect." + "reason": "Both involve configs, but the specific discrepancies and affected models differ; not the same bug." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Video-processing ValueError in Qwen3OmniMoe and batched apply_chat_template crash in Qwen2_5_VLProcessor are different processor code paths." + "reason": "Both are multimodal processor bugs, but one is Qwen3OmniMoe video processing and the other is batched apply_chat_template padding behavior in Qwen2_5_VL." }, { - "left": "issue:43673", - "right": "issue:44743", + "left": "issue:42175", + "right": "issue:43316", "accept": false, - "reason": "GenerationMixin cache removal/chunked_prefill issue versus recurrent-state reset in modular_qwen3_5; both cache-related but not the same bug." + "reason": "Packaging/backend dependency omission is unrelated to a Gemma3TextConfig API discrepancy." }, { - "left": "issue:42175", - "right": "issue:43316", + "left": "issue:43673", + "right": "issue:44743", "accept": false, - "reason": "Package/backend installation coverage issue versus a config-class API discrepancy; no common underlying bug." + "reason": "Both mention cache, but the first is missing GenerationMixin cache in chunked_prefill while the second is recurrent state reset in modular qwen3_5; different code paths." }, { - "left": "issue:43404", - "right": "issue:43441", + "left": "issue:43065", + "right": "issue:44617", "accept": false, - "reason": "Mistral3 weight-tying bug and Ministral-3 FlashAttention failure are distinct model/runtime problems." + "reason": "A dummy Conv2d implementation issue is not the same as a CUDA OOM in Sam3Video." }, { - "left": "issue:43065", - "right": "issue:44617", + "left": "issue:43404", + "right": "issue:43441", "accept": false, - "reason": "Dummy Conv2d implementation in Sam3PixelDecoder and Sam3Video CUDA OOM are not the same code-path problem." + "reason": "Mistral3 lm_head tying and Ministral-3 FlashAttention failure are distinct bugs." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 model breakage and janus image-generation test failure are unrelated." + "reason": "ESM2 breakage affecting workflows and a janus image-generation test failure are unrelated." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "Custom processor.tokenizer regression and pipeline() error-message/docs issue are separate pipeline-related but non-identical changes." + "reason": "Processor.tokenizer/image passing regression is different from a pipeline error message about translation support." }, { "left": "issue:25251", "right": "issue:28282", "accept": false, - "reason": "Pipeline top_k nesting bug and missing PyTorch import error are unrelated." + "reason": "Pipeline top_k nesting behavior is unrelated to a missing PyTorch dependency ImportError." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Metric gathering label truncation and auxiliary-loss normalization in MoE models are different bugs in different parts of the stack." + "reason": "Label truncation in gather_for_metrics is unrelated to auxiliary-loss normalization in OLMoE/GPT Oss." } ] }, @@ -9160,8 +9076,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -9218,8 +9134,8 @@ "issue:44077|issue:44534", "issue:16998|issue:22355", "issue:43550|issue:43701", - "issue:43761|issue:44479", "issue:44871|issue:45584", + "issue:43761|issue:44479", "issue:43606|issue:43701", "issue:45127|issue:45200", "issue:43295|issue:44871", @@ -9230,8 +9146,8 @@ "issue:43646|issue:44913", "issue:43504|issue:45698", "issue:44617|issue:45584", - "issue:43931|issue:44060", "issue:43761|issue:44242", + "issue:43931|issue:44060", "issue:45563|issue:45584", "issue:32090|issue:35141", "issue:44464|issue:44871", @@ -9245,226 +9161,214 @@ "issue:44534|issue:44913", "issue:44190|issue:44964" ], - "prepared_review_unit_hash": "dd4e9a6e109718e7217b5c1a03e6e21f75ce5e4eb9de5fa32c8936d2e4933fc6", + "prepared_review_unit_hash": "2e4e383b1b7301f3b6a9c5a6cbf4247fa6bce2ad146eaaefa5fca5a9948f2d87", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36064, - "estimated_input_tokens": 9016, - "estimated_eval_tokens": 18288 + "serialized_chars": 36073, + "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 mostly a grab bag of unrelated Transformers issues: model loading/config regressions, multimodal pipeline bugs, generation edge cases, and docs/runtime problems. None of the soft pairs look like the same concrete underlying bug/change, so I would not merge any of them as duplicates.", - "confidence": 0.93, - "canonical_issue_reason": "Issue 44521 is the best anchor because it is open, has the most inbound references, and describes a broadly impactful multimodal chat-template bug with clear reproduction details.", + "summary": "The set is heterogeneous: most soft pairs share only broad subsystem or wording similarity, not the same underlying bug. No pull requests are present. Issue 44521 is the best anchor only because it is open, active, and well-scoped.", + "confidence": 0.42, + "canonical_issue_reason": "Issue 44521 is the strongest canonical anchor in this set: it is open, has the most inbound references, and describes a concrete reproducible multimodal chat-template regression.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44521 is the most suitable representative issue in this set due to its higher visibility/activity and a well-scoped, user-facing regression description.", + "best_issue_reason": "Issue 44521 is the most triage-worthy representative issue because it is active, well-described, and likely to attract follow-up reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Different bugs and code paths: video-classification processor lookup vs SAM3 vision encoder attribute access." + "reason": "Different bugs: pipeline processor lookup vs missing model output attribute in SAM3 video." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "Unrelated: patchtsmixer post_init policy vs non-persistent buffer initialization regression." + "reason": "One is a post_init allowance policy change; the other is a v5 buffer initialization regression." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "Completely different topics: model_max_length behavior vs missing transformers.onnx module." + "reason": "Unrelated topics: DeBERTa model_max_length question vs missing transformers.onnx module." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "Different failures: torch.compile/SDPA on Bamba vs resume_from_checkpoint key mismatch." + "reason": "SDPA/torch.compile failure is unrelated to resume_from_checkpoint key mismatch." }, { - "left": "issue:43761", - "right": "issue:44479", + "left": "issue:44871", + "right": "issue:45584", "accept": false, - "reason": "Different models and symptoms: CLIPVision hidden_states regression vs Qwen VL video input regression." + "reason": "Different code paths: eos_token_id config mismatch vs Whisper empty-transcription generation failure." }, { - "left": "issue:44871", - "right": "issue:45584", + "left": "issue:43761", + "right": "issue:44479", "accept": false, - "reason": "Both involve generation/tokenization, but the concrete failures differ: eos_token_id mismatch vs empty transcription after align_special_tokens." + "reason": "Both are regressions, but on different models and different failures; not the same bug." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "Device mismatch with CPU offload is unrelated to checkpoint key mismatch on resume." + "reason": "CPU offload device mismatch is unrelated to checkpoint key mismatch." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "Different model behaviors: LoRA merge with extended vocab vs Gemma 4 multimodal token type ids defaulting." + "reason": "LoRA merge / tied embeddings issue is different from Gemma 4 mm_token_type_ids defaults." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Different regression surfaces: processor/tokenizer API breakage vs Gemma eos token configuration mismatch." + "reason": "Processor/tokenizer regression is unrelated to eos_token_id configuration drift." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "Both are loading issues, but one is gguf_file/from_pretrained behavior and the other is missing model_type in config." + "reason": "Both concern loading, but the root causes differ: gguf_file handling vs missing model_type in config." }, { "left": "issue:42617", "right": "issue:43575", "accept": false, - "reason": "Different problems: running 3d_parallel.py vs tensor-parallel OOM when loading a large model." + "reason": "A script startup failure and an OOM during tensor-parallel loading are not the same defect." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "OOM during TP loading is unrelated to a key error on MobileLLM load." + "reason": "OOM during model load does not match a key error for MobileLLM loading." }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "Different serve failures: KV cache continuation indexing vs Gemma4Processor missing _tokenizer." + "reason": "Both hit transformers serve, but one is KV-cache indexing and the other is a missing processor attribute." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Custom model initialization breakage is unrelated to GPTNeoXConfig rotary_pct not persisting on reload." + "reason": "Custom model initialization breakage is not the same as GPTNeoXConfig reload reverting rotary_pct." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "Different loading bugs: legacy preset field handling vs wrong custom module loaded after save_pretrained." + "reason": "Legacy field handling in BEiT preset loading is unrelated to wrong custom-module selection after save_pretrained." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "CUDA OOM in Sam3Video is unrelated to Whisper generation failure on empty transcription." + "reason": "CUDA OOM in Sam3Video is unrelated to Whisper empty transcription behavior." }, { - "left": "issue:43931", - "right": "issue:44060", + "left": "issue:43761", + "right": "issue:44242", "accept": false, - "reason": "Shape mismatch when loading a model is not the same bug as tied-weights warnings wiring the wrong tensors." + "reason": "Hidden_states missing in CLIPVisionModel is different from missing load-balancing loss when router logits are off." }, { - "left": "issue:43761", - "right": "issue:44242", + "left": "issue:43931", + "right": "issue:44060", "accept": false, - "reason": "Hidden_states=None regression in CLIPVisionModel is unrelated to missing load-balancing loss when output_router_logits=False." + "reason": "Shape mismatch on Qwen3-VL load is not the same as tied-weights warning in Qwen3-Next." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "Both are generation-related, but stale warnings and empty transcription crashes are different issues." + "reason": "A stale warning in paged generate is unrelated to a generation crash on empty Whisper output." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer GPU broadcast NoneType error is unrelated to embedding reinitialization during post_init." + "reason": "Trainer broadcast NoneType error is unrelated to embedding reinitialization after resize." }, { "left": "issue:44464", "right": "issue:44871", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is a different code path from eos_token_id config mismatch." + "reason": "Chunked generation inconsistency with compiled forward is unrelated to eos_token_id config inconsistency." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Video input regression for Qwen VL models is unrelated to serve KV-cache continuation indexing." + "reason": "Both are regressions, but one concerns video input processing and the other server-side KV-cache indexing." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "SAM3 pixel decoder dummy conv and OmDet-Turbo processor image size mismatch are different model-specific bugs." + "reason": "Dummy Conv2d in Sam3PixelDecoder is different from OmDet-Turbo processor/model input-size mismatch." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Janus image-generation test failure is unrelated to Qwen3.5 flash-attention generation behavior." + "reason": "Failed Janus image-generation test is unrelated to Qwen3.5 flash-attention generation." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "Both touch multimodal chat handling, but one is assistant mask generation and the other is video vision_position_ids; not the same bug." + "reason": "Both are multimodal, but assistant mask generation and video vision_position_ids are different bugs." }, { "left": "issue:43906", "right": "issue:45290", "accept": false, - "reason": "Isolated reproduction of another issue is not the same as crashes on assistant messages with tool calls and no content." + "reason": "An isolated reproduction of an older issue is not enough to match the tool-call/no-content chat-template crash." }, { "left": "issue:43827", "right": "issue:43994", "accept": false, - "reason": "Docs still referencing pipeline() is unrelated to SigLIP2 producing nonsensical results with AutoModel/pipeline." + "reason": "Documentation references to removed pipelines do not match a model-output correctness problem." }, { - "left": "issue:44534", - "right": "issue:44913", - "accept": false, - "reason": "Non-persistent buffer junk initialization is unrelated to rotary_pct not surviving config reload." - }, - { - "left": "issue:44190", - "right": "issue:44964", + "left": "issue:44464", + "right": "issue:45563", "accept": false, - "reason": "Local dataset loading in a script is unrelated to Phi-4 multimodal model loading failure." + "reason": "Generation inconsistency and a stale warning are different severity and different failure modes." }, { - "left": "issue:43452", - "right": "issue:45698", + "left": "issue:44534", + "right": "issue:44913", "accept": false, - "reason": "Different from_pretrained/save_pretrained issues: gguf_file handling vs wrong custom module resolution." + "reason": "Non-persistent buffer junk in v5 is unrelated to GPTNeoX config reload defaults." }, { - "left": "issue:43931", + "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Weight shape mismatch for a specific model is not the same as an overall latest-transformers load failure for Phi-4 multimodal." - }, - { - "left": "issue:44464", - "right": "issue:45563", - "accept": false, - "reason": "Compiled chunked generation inconsistency is different from a stale warning about num_return_sequences." + "reason": "Local dataset loading in an example script is unrelated to loading a multimodal model preset." } ] }, @@ -9483,8 +9387,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -9554,9 +9458,9 @@ "issue:43644|issue:44913", "issue:43540|issue:44734", "issue:44112|issue:45600", - "issue:43650|issue:44220", "issue:44496|issue:45803", "issue:43335|issue:44190", + "issue:43650|issue:44220", "issue:45276|issue:45584", "issue:39692|issue:44190", "issue:45245|issue:45584", @@ -9568,214 +9472,214 @@ "issue:44186|issue:44734", "issue:38175|issue:43065" ], - "prepared_review_unit_hash": "bb8217c596c506a487327ef573b22fe71885f2508932b3b036303e284caa2472", + "prepared_review_unit_hash": "a5a112a68f564965a0408cf3f9900ba9ba3db224c9f535914556750a49544df1", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36320, - "estimated_input_tokens": 9080, - "estimated_eval_tokens": 18416 + "serialized_chars": 36323, + "estimated_input_tokens": 9081, + "estimated_eval_tokens": 18418 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the items cover unrelated import errors, model/config bugs, runtime failures, and docs/CI issues. None of the soft pairs look like the same underlying bug or change, so all soft edges should be rejected.", - "confidence": 0.98, - "canonical_issue_reason": "No clear canonical issue: the set does not form a single duplicate topic, and the closest title overlap is still only a loose token-embedding/config family among many unrelated bugs.", + "summary": "The soft pairs are mostly high-similarity false positives: they overlap in broad subsystem vocabulary (imports, configs, pipelines, CI, model loading) but describe different models, code paths, and failure modes. I would not deduplicate any of them.", + "confidence": 0.97, + "canonical_issue_reason": "If a single issue must be chosen as the representative, issue 45276 is the clearest, most concrete standalone bug in this set: it precisely describes a resize-token-embeddings/output-embeddings mismatch with a reproducible model behavior change.", "canonical_pr_reason": null, - "best_issue_reason": "No strong global best issue; the cluster is too mixed for a single representative duplicate target.", + "best_issue_reason": "Issue 45276 is the strongest issue candidate because it is specific, actionable, and tied to a concrete state inconsistency after embedding resizing; it is the most understandable representative among otherwise unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "Different problems: missing image_utils import vs a Qwen3.5 chat crash." + "reason": "Different problems: an invalid import from image_utils vs a Qwen3.5 chat runtime crash." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Unrelated failures: CUDA device-mapping assert vs category-count limit." + "reason": "Unrelated failures: auto device mapping CUDA assert vs category-count limit error." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "One is a janus image-generation test failure; the other is a broken auto_mappings reference/CI issue." + "reason": "Both are CI-ish, but one is a janus image-generation test failure and the other is a stale auto-mapping reference bug." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Python 3.13 import parsing error vs flash-attention crash on sink-less models." + "reason": "Completely different code paths: Python 3.13 parsing/IndentationError vs flash-attention aux-state handling." }, { "left": "issue:44737", "right": "issue:45803", "accept": false, - "reason": "CPU device placement bug in XLNet vs bare except swallowing errors in Fuyu batching." + "reason": "Different bugs: XLNet device placement in positional encoding vs a bare-except swallowing errors in Fuyu tensor conversion." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "LoRA/vocab merge collapse vs incorrect visual encoder keys on save_pretrained." + "reason": "Same model family, but distinct issues: LoRA merge with extended vocab vs incorrect visual encoder keys in save_pretrained." }, { "left": "issue:43388", "right": "issue:44079", "accept": false, - "reason": "Metric gathering drops tuple labels, while the other issue is about ModelOutput key assignment." + "reason": "One is metric gathering dropping tuple labels; the other is ModelOutput field assignment with None keys." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Token-embedding resize bug vs path-with-slash import warning; no shared code path." + "reason": "Embedding reinitialization after resize is unrelated to the local-path trailing-slash import warning." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json vs list_repo_files not forwarding revision." + "reason": "Both involve repository/file lookup, but tokenizer.json precedence and list_repo_files revision forwarding are different APIs and failures." }, { "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "SwitchTransformers sparse-layer construction bug vs DebertaV2 import failure on Python 3.13." + "reason": "Different models and bugs: SwitchTransformers sparse-layer creation vs DebertaV2 Python 3.13 import parsing." }, { "left": "issue:44077", "right": "issue:44079", "accept": false, - "reason": "post_init policy issue vs ModelOutput None-key assignment bug." + "reason": "Unrelated: patchtsmixer post_init policy vs ModelOutput key reassignment behavior." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 loading failure vs missing TensorFlow backend handling." + "reason": "Python 3.14 import/load failure is not the same as the tensorboard-without-TensorFlow backend error." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "torch.compile/SDPA model failure vs dataset-loading problem in a training script." + "reason": "Different surfaces entirely: torch.compile/SDPA failure in Bamba vs dataset loading in an image-classification script." }, { "left": "issue:43976", "right": "issue:44855", "accept": false, - "reason": "Python-version compatibility issue vs DebertaV2 parsing/import problem." + "reason": "Both are version-compatibility reports, but one targets Python 3.9/3.10 behavior and the other is a Python 3.13 IndentationError." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer initialization vs GPTNeoX config reload losing rotary_pct." + "reason": "Non-persistent buffer initialization junk is not the same as GPTNeoX rotary_pct not persisting on reload." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-input validation vs serve KV-cache continuation indexing." + "reason": "Different runtime paths: Qwen3OmniMoe video-input validation vs transformers serve KV-cache indexing." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "GraniteSpeech CI test flake vs removed Sam3LiteText config references." - }, - { - "left": "issue:43650", - "right": "issue:44220", - "accept": false, - "reason": "Placeholder issue vs fbank feature extraction bug." + "reason": "A stale GraniteSpeech test is not the same as broken auto_mappings references to removed configs." }, { "left": "issue:44496", "right": "issue:45803", "accept": false, - "reason": "Unknown model_type loading failure vs bare except swallowing interrupts/errors." + "reason": "Model config discovery/import failure is unrelated to FuyuBatchFeature's broad exception handling." }, { "left": "issue:43335", "right": "issue:44190", "accept": false, - "reason": "SwitchTransformers config construction bug vs local dataset loading failure." + "reason": "Different subsystems and symptoms: SwitchTransformers config creation vs local dataset loading in a training script." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "The placeholder 'ADD THE DATA' issue is not the same as an audio feature extraction bug." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Gemma4 resize_token_embeddings/output_embeddings bug vs Whisper empty-transcription generation failure." + "reason": "Embedding-resize/output-embedding inconsistency is unrelated to Whisper empty-transcription generation after token alignment." }, { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "SigLIP2 docs example errors vs local dataset loading in an image-classification script." + "reason": "Docs example/quantization failures are distinct from a local dataset loading problem." }, { "left": "issue:45245", "right": "issue:45584", "accept": false, - "reason": "Category-count runtime limit vs Whisper generation bug." + "reason": "Different errors and models: category cardinality limit vs Whisper generation on empty transcription." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Video-classification pipeline processor mismatch vs notebook syntax error." + "reason": "One is a video-classification pipeline bug; the other is a notebook syntax error." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Phi4Multimodal default config handling vs missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig." + "reason": "Both are config issues, but they affect different multimodal configs and different fields/defaults." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Processor/tokenizer regression vs Gemma4 token-embedding resize behavior." + "reason": "Regression in processor.tokenizer/image passing is unrelated to Gemma4 token-embedding resize behavior." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Specific Qwen2.5 VL image-processor import failure vs AutoImageProcessor import failure." + "reason": "Both are import errors, but they involve different symbols and different model modules." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Processor import error vs configuration_utils symbol import error." + "reason": "Different missing imports from different modules; not the same underlying bug." }, { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "LayoutLMv2 tokenizer padding/NER crash vs serve KV-cache tensor indexing." + "reason": "Tokenizer padding/truncation crash and serve KV-cache indexing crash are separate failures." }, { "left": "issue:38175", "right": "issue:43065", "accept": false, - "reason": "SigLIP2 zero-probabilities bug vs Sam3PixelDecoder dummy Conv2d issue." + "reason": "Different models and mechanisms: SigLIP2 zero probabilities vs Sam3PixelDecoder dummy Conv2d." } ] }, @@ -9794,8 +9698,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -9849,8 +9753,8 @@ ], "soft_pairs": [ "issue:44683|issue:45446", - "issue:44016|issue:44220", "issue:42175|issue:44279", + "issue:44016|issue:44220", "issue:44466|issue:44913", "issue:44734|issue:45632", "issue:43404|issue:45216", @@ -9866,234 +9770,234 @@ "issue:43976|issue:44987", "issue:43295|issue:45335", "issue:43673|issue:44336", - "issue:43824|issue:44933", "issue:43531|issue:45600", + "issue:43824|issue:44933", "issue:44496|issue:45042", "issue:43526|issue:43901", "issue:43299|issue:45216", "issue:44075|issue:44368", "issue:45335|issue:45584", "issue:43976|issue:44279", - "issue:43299|issue:45357", "issue:43381|issue:45593", + "issue:43299|issue:45357", "issue:43901|issue:44496", - "issue:44351|issue:45278", - "issue:44496|issue:45759" + "issue:44496|issue:45759", + "issue:44351|issue:45278" ], - "prepared_review_unit_hash": "bba8e89f0bc52adfa0006d0cc18b78eb29f797502c0fe1f04fcc470261c1408e", + "prepared_review_unit_hash": "0b8816fae2909264dc6cae672aa096cccdca1187a24b84a208a3bc33b3bcc904", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 36355, - "estimated_input_tokens": 9089, - "estimated_eval_tokens": 18434 + "serialized_chars": 36350, + "estimated_input_tokens": 9088, + "estimated_eval_tokens": 18432 }, "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 bugs, regressions, and import/load issues. None of the proposed soft pairs look like the same concrete underlying bug or change, so all soft edges are rejected.", - "confidence": 0.45, - "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the list ('Many import errors after update from 4.57.0 to 5.5.0') and is the closest thing to a cluster representative.", + "summary": "These items are a heterogeneous set of unrelated Transformers issues; the soft pairs mostly only share broad subsystem or regression language, not the same concrete bug. No soft edge should be merged.", + "confidence": 0.97, + "canonical_issue_reason": "No true duplicate cluster emerges from these candidates. If one issue must serve as an umbrella, issue 45278 is the broadest (many import errors after upgrade), but it is still not the same underlying bug as the others.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best single representative because it captures the widest import-regression surface among these issues, though the overall set is not a true duplicate cluster.", + "best_issue_reason": "issue 45278 is the most general and umbrella-like issue in the set, so it is the least bad anchor for triage, though the set is too diverse to treat as one duplicate cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Both mention flex_attention/PyTorch, but one is a compiled-kernel failure on torch>=2.9 while the other is a version-check/import bug for AuxRequest. Different concrete failures." + "reason": "Different failures: compiled flex_attention on torch>=2.9 vs a wrong PyTorch version check for AuxRequest import." }, { - "left": "issue:44016", - "right": "issue:44220", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "A notebook syntax error and a feature-extraction bug are unrelated problems in different code paths." + "reason": "Packaging dependency problems, but one is missing TensorFlow in extras and the other is a general transformers dependency issue; not the same bug." }, { - "left": "issue:42175", - "right": "issue:44279", + "left": "issue:44016", + "right": "issue:44220", "accept": false, - "reason": "One is about backend extras not pulling in TensorFlow; the other is a vague dependency issue. Too different to be the same bug." + "reason": "Notebook syntax error vs a feature-specific audio feature extraction issue; unrelated code paths." }, { "left": "issue:44466", "right": "issue:44913", "accept": false, - "reason": "Both involve model config/serialization behavior, but one is tied lm_head serialization and the other is rotary_pct reload loss. Different fields and code paths." + "reason": "Both involve config/serialization behavior, but one is tied lm_head.weight saving and the other is rotary_pct not persisting on reload." }, { "left": "issue:44734", "right": "issue:45632", "accept": false, - "reason": "One is a server KV-cache indexing crash; the other is a trust_remote_code cache-path collision for local models. Not the same issue." + "reason": "Serve/KV-cache tensor indexing crash is unrelated to a trust_remote_code cache-path collision for local models." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Both are tied-weights/save-pretrained regressions, but they affect different models and symptoms. Not enough to be the same concrete bug." + "reason": "Weight tying in Mistral3 conditional generation is unrelated to Qwen3.5 save_pretrained checkpoint corruption." }, { "left": "issue:44561", "right": "issue:45632", "accept": false, - "reason": "Removal of is_torch_fx_available breaking trust_remote_code models is unrelated to local cache-path collisions." + "reason": "Removal of a helper breaking trust_remote_code models is a different failure mode from local cache-path collisions." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import-time slowness and unsafe sys.modules access may be loosely related, but they describe different bugs and fixes." + "reason": "Import latency during transformers import is not the same as unsafe sys.modules access in modeling_utils." }, { "left": "issue:42490", "right": "issue:45632", "accept": false, - "reason": "Both involve local model paths, but one is a trailing-slash warning and the other is a leaf-directory cache collision. Different failure modes." + "reason": "Local path ending with slash import warning is different from cache path collisions between same leaf directory names." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "One is missing fields when loading remote-code configs; the other is an unrecognized model/config.json model_type issue." + "reason": "Missing fields from trust_remote_code AutoConfig return values is not the same as an unrecognized-model/config.json model_type error." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "Loading Phi-4 multimodal and stale auto_mappings references to removed Sam3LiteText configs are unrelated." + "reason": "Model loading failure for Phi-4 multimodal is unrelated to stale auto_mappings references to removed Sam3LiteText configs." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "A CI test fixture/device override problem is unrelated to a model loading regression." + "reason": "A CI test regression on GraniteSpeech device override is unrelated to loading Phi-4 multimodal models." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Both are Ministral-family issues, but one is about saving dequantized weights and the other is FlashAttention loading. Different bug classes." + "reason": "Saving dequantized Ministral/Devstral models and FlashAttention loading failures are separate model-specific bugs." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "Different models and different load-path failures; no concrete shared bug." + "reason": "Unit-test load failure for GLM-4.7-Flash and Phi-4 multimodal loading failure are different models and code paths." }, { "left": "issue:44016", "right": "issue:44062", "accept": false, - "reason": "Notebook syntax error versus AddedToken keyword-argument conflict are unrelated." + "reason": "Notebook syntax error and AddedToken keyword collision are unrelated issues." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "A Python-version compatibility problem and a model-loading regression are not the same underlying bug." + "reason": "Python version compatibility breakage is not the same as a specific model loading failure on 5.1.0+." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "These are both model-specific regressions, but one is about processor/tokenizer access and image passing, while the other is about resize_token_embeddings on t5gemma." + "reason": "Processor/tokenizer regression and t5gemma resize_token_embeddings bug affect different APIs and model components." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "A GenerationMixin cache regression and ANSI escape handling in loading_report are unrelated." + "reason": "Generation cache missing during chunked_prefill is unrelated to ANSI codes emitted by loading_report." }, { - "left": "issue:43824", - "right": "issue:44933", + "left": "issue:43531", + "right": "issue:45600", "accept": false, - "reason": "Both are import-related, but they concern different symbols and different export failures." + "reason": "Qwen3-MoE sliding_window behavior and stale auto_mappings references are different bugs." }, { - "left": "issue:43531", - "right": "issue:45600", + "left": "issue:43824", + "right": "issue:44933", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to stale config references in auto_mappings.py." + "reason": "Qwen2.5-VL import/export failure is not the same as a nonexistent import from image_utils." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "An unrecognized model/config error is unrelated to PIL image processors incorrectly requiring torchvision." + "reason": "Unrecognized model/config error is different from PIL image processors incorrectly requiring torchvision." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "A BEiT image processor label reduction bug and a docs mismatch in TextClassificationPipeline are different issues." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to TextClassificationPipeline docs mismatch." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Both involve Qwen3.x regressions, but one is model loading and the other is incorrect checkpoint serialization. Different code paths." + "reason": "Qwen3VL MoE loading breakage is a different model family and failure mode from Qwen3.5 save_pretrained corruption." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Unused SGD args and a tie_word_embeddings warning during LoRA fine-tuning are unrelated." + "reason": "SGD optimizer arguments not being used is unrelated to a tie_word_embeddings warning in Qwen3.5 fine-tuning." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "A resize_token_embeddings bug in t5gemma and Whisper empty-transcription generation failure are different model behaviors." + "reason": "t5gemma embedding resize bug is unrelated to Whisper empty transcription generation after token alignment." }, { "left": "issue:43976", "right": "issue:44279", "accept": false, - "reason": "A version/platform compatibility complaint and a generic dependency issue are too broad and not the same defect." + "reason": "Python compatibility and dependency issues are both environment-related, but they are not the same concrete failure." }, { - "left": "issue:43299", - "right": "issue:45357", + "left": "issue:43381", + "right": "issue:45593", "accept": false, - "reason": "Both are Qwen3.5 regressions, but one breaks model loading while the other saves incorrect visual encoder keys. Distinct bugs." + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary losses with denoising off are unrelated training behaviors." }, { - "left": "issue:43381", - "right": "issue:45593", + "left": "issue:43299", + "right": "issue:45357", "accept": false, - "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are unrelated training-time issues." + "reason": "Both are Qwen3.5-related regressions, but one is loading failure and the other is incorrect save_pretrained visual encoder keys." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Docs/behavior mismatch for text classification is unrelated to an unrecognized-model config error." + "reason": "Docs mentioning return_all_scores and a model config loading error are not the same bug." }, { - "left": "issue:44351", - "right": "issue:45278", + "left": "issue:44496", + "right": "issue:45759", "accept": false, - "reason": "A missing HybridCache export is one specific import error, while 45278 is a broad umbrella report; not the same concrete bug." + "reason": "Unrecognized model config loading failure is different from AutoModelForCausalLM not unwrapping text_config for composite multimodal configs." }, { - "left": "issue:44496", - "right": "issue:45759", + "left": "issue:44351", + "right": "issue:45278", "accept": false, - "reason": "Missing model_type in config.json and unwrapping text_config for multimodal configs are different config-loading problems." + "reason": "Missing HybridCache import is a specific symbol export issue, not the broad import-error cluster described by 45278." } ] }, @@ -10112,8 +10016,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -10174,8 +10078,8 @@ "issue:43479|issue:43525", "issue:44779|issue:45812", "issue:41950|issue:45823", - "issue:43577|issue:45216", "issue:38175|issue:44617", + "issue:43577|issue:45216", "issue:44683|issue:45542", "issue:39692|issue:43901", "issue:43299|issue:43854", @@ -10187,218 +10091,211 @@ "issue:45081|issue:45907", "issue:44016|issue:44291", "issue:43441|issue:45834", - "issue:45600|issue:45712", "issue:33290|issue:45656", + "issue:45600|issue:45712", "issue:43450|issue:44279", "issue:43065|issue:44898", "issue:42994|issue:43317", "issue:44466|issue:45276", "issue:44261|issue:44263", - "issue:43650|issue:45290", - "issue:43441|issue:44683" + "issue:43650|issue:45290" ], - "prepared_review_unit_hash": "bad89ae20be48b1773be4561649440b3a8ba384efe5a67301cbdbf196703eb53", + "prepared_review_unit_hash": "b49b549da53e24330d425c68000b5931dc6369d387d8d5fa41c09812b82d64cf", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36421, - "estimated_input_tokens": 9106, - "estimated_eval_tokens": 18468 + "soft_pair_count": 28, + "serialized_chars": 36117, + "estimated_input_tokens": 9030, + "estimated_eval_tokens": 18316 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated issue reports across different models and subsystems. The suggested soft matches are only broad topical similarities (tokenizers, multimodal processing, deepspeed, loading, serialization) and do not look like the same underlying bug, so none should be merged.", - "confidence": 0.94, - "canonical_issue_reason": null, + "summary": "This cluster is largely a collection of unrelated Transformers issues with superficial similarity around regressions, tokenizers, multimodal models, and deepspeed. I found no strong duplicate pairs to merge; the soft matches are mostly same-subsystem, not same-bug matches.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45020 is the broadest representative report here: it describes a recent-version regression affecting model loading/remote code and best matches the cluster\u2019s overall \u2018v5 breakage\u2019 theme.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:45020 is the most suitable anchor because it is broad, open, and discussion-heavy, while the other issues are narrower, model-specific failures.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44521", "right": "issue:45325", "accept": false, - "reason": "Both touch multimodal behavior, but one is assistant masks in chat templating and the other is RoPE temporal \u0564\u056b\u0580 position scaling; different code paths and symptoms." + "reason": "Both are multimodal, but one is a chat-template masking bug and the other is a RoPE/position-id scaling issue for Qwen2.5-VL. Different code paths and symptoms." }, { "left": "issue:41950", "right": "issue:45020", "accept": false, - "reason": "A video-classification processor lookup bug is not the same as a broad regression affecting remote_code models." + "reason": "Video-classification processor lookup and remote-code model breakage are unrelated failures." }, { "left": "issue:43976", "right": "issue:45405", "accept": false, - "reason": "These are separate version-compatibility problems: Python support vs a PEFT minimum-version bump." + "reason": "Python version support failure and a PEFT minimum-version bump are not the same bug." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Different bugs in different model families: missing load-balancing loss vs Gemma4 mm_token_type_ids defaults." + "reason": "Load-balancing loss gating and Gemma4 mm_token_type_ids defaults are different model behaviors." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "One is default multimodal config initialization, the other is a missing pad_token_id attribute; not the same defect." + "reason": "Different configs and different failures: default multimodal config initialization vs missing pad_token_id on Llama4Config." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different model families and likely different tokenizer logic." + "reason": "Both are tokenizer regressions, but they target different model families and likely different tokenizer code paths." }, { "left": "issue:41950", "right": "issue:45823", "accept": false, - "reason": "Processor lookup in a video pipeline is unrelated to a Gemma4 device-mapping forward mismatch." + "reason": "Processor selection for video classification is unrelated to Gemma4 device placement during forward." }, { - "left": "issue:43577", - "right": "issue:45216", + "left": "issue:38175", + "right": "issue:44617", "accept": false, - "reason": "One is dtype propagation on load, the other is incorrect save_pretrained serialization; different lifecycle stages and failures." + "reason": "Zero probabilities in SigLIP2 and CUDA OOM in Sam3Video are unrelated issues." }, { - "left": "issue:38175", - "right": "issue:44617", + "left": "issue:43577", + "right": "issue:45216", "accept": false, - "reason": "Zero probabilities in a model output is unrelated to CUDA OOM in Sam3Video." + "reason": "BLIP2 dtype propagation and Qwen3.5 save_pretrained serialization are distinct bugs." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "A torch flex_attention regression is unrelated to a tensorboard/TensorFlow backend installation error." + "reason": "Compiled flex_attention on torch 2.9 and tensorboard/TensorFlow backend installation are unrelated." }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "Both mention docs, but one is a specific SigLIP2 example failure and the other is a pipeline option documentation mismatch." + "reason": "Both involve docs, but one is a SigLIP2 example error and the other is a pipeline docs mismatch." }, { "left": "issue:43299", "right": "issue:43854", "accept": false, - "reason": "Different model-loading failures for different models; no shared concrete code path is evident." + "reason": "Different model families and different loading failures; not the same underlying regression." }, { "left": "issue:44279", "right": "issue:45405", "accept": false, - "reason": "A generic dependency complaint does not match a specific PEFT version release issue." + "reason": "A generic dependency issue is not the same as a specific unreleased PEFT version bump." }, { "left": "issue:41950", "right": "issue:43650", "accept": false, - "reason": "The titles are unrelated; one is a pipeline processor bug, the other is a non-descriptive report." + "reason": "Unrelated: video pipeline processor lookup vs an unrelated placeholder issue." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Processor/tokenizer API regression and DeepSeek tokenization errors are distinct tokenizer-related bugs." + "reason": "Processor/tokenizer regression in v4.57.5 is separate from Deepseek tokenizer misbehavior in v5." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "Ignoring tokenizer.json and crashing in Mistral regex patching are different tokenizer-loading problems." + "reason": "AutoTokenizer ignoring tokenizer.json and Mistral regex patch crashing are different tokenizer-loading bugs." }, { "left": "issue:44297", "right": "issue:44462", "accept": false, - "reason": "Tokenizer class metadata mismatch on save_pretrained is not the same as AutoTokenizer ignoring repository tokenizer.json." + "reason": "Tokenizer class serialization mismatch is not the same as repository tokenizer.json being ignored." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Tokenizer patching crash and list_repo_files revision forwarding are entirely separate utilities." + "reason": "Mistral tokenizer loading and list_repo_files revision forwarding are unrelated." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax errors and init_empty_weights argument errors are unrelated." + "reason": "Notebook syntax error and init_empty_weights argument mismatch are unrelated." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "FlashAttention failure on one model and OCR long-input index error on another are different bugs." + "reason": "FlashAttention failure on Ministral-3 and OCR index error in Kosmos2.5 are different model-specific bugs." }, { - "left": "issue:45600", - "right": "issue:45712", + "left": "issue:33290", + "right": "issue:45656", "accept": false, - "reason": "Removed config references in auto_mappings and leftover dummy classes are separate repo-maintenance issues." + "reason": "Both mention deepspeed/optimizers, but OOM with Adafactor and double optimizer.step calls are different issues." }, { - "left": "issue:33290", - "right": "issue:45656", + "left": "issue:45600", + "right": "issue:45712", "accept": false, - "reason": "Both involve deepspeed/optimizers, but one is OOM with Adafactor and the other is duplicate optimizer stepping; not the same issue." + "reason": "Both are repo hygiene/CI problems, but they concern different files and failure modes." }, { "left": "issue:43450", "right": "issue:44279", "accept": false, - "reason": "A batched video-processor shape bug is unrelated to a generic dependency issue." + "reason": "Video processor batching shape bug and a dependency issue are unrelated." }, { "left": "issue:43065", "right": "issue:44898", "accept": false, - "reason": "A dummy Conv2d placeholder in Sam3PixelDecoder is not the same as a Perceiver interpolation failure." + "reason": "Different vision-model bugs: dummy Conv2d in Sam3PixelDecoder vs Perceiver non-default resolution failure." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Saving quantized models and loading dequantized models with device_map offload are different quantization pathways." + "reason": "Quantized model saving failure and dequantized load/offload failure are distinct stages and bugs." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Inconsistent lm_head serialization and resize_token_embeddings not updating Gemma4 embeddings are separate weight-handling bugs." + "reason": "Serialization of lm_head.weight and resize_token_embeddings not updating Gemma4 outputs are separate problems." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "Both are GLM-related, but one is an rms_norm_eps/config precision issue and the other is a torch.split return-value issue." + "reason": "MLA layernorm epsilon mismatch and GlmMoeDsaIndexer split-return handling are unrelated." }, { "left": "issue:43650", "right": "issue:45290", "accept": false, - "reason": "A vague issue title and a concrete apply_chat_template crash do not indicate the same bug." - }, - { - "left": "issue:43441", - "right": "issue:44683", - "accept": false, - "reason": "FlashAttention failures and compiled flex_attention failures are different attention-stack regressions, not one bug." + "reason": "Placeholder issue and apply_chat_template crash with tool-call assistant messages are unrelated." } ] }, @@ -10417,19 +10314,19 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ "issue:41628", - "issue:41950", "issue:42175", "issue:42491", "issue:43122", "issue:43262", - "issue:43295", "issue:43299", + "issue:43441", + "issue:43450", "issue:43525", "issue:43643", "issue:43650", @@ -10446,7 +10343,6 @@ "issue:44016", "issue:44162", "issue:44188", - "issue:44220", "issue:44242", "issue:44246", "issue:44295", @@ -10454,12 +10350,12 @@ "issue:44336", "issue:44361", "issue:44373", - "issue:44496", "issue:44521", "issue:44589", "issue:44617", "issue:44623", "issue:44671", + "issue:44683", "issue:44749", "issue:44779", "issue:44987", @@ -10473,14 +10369,15 @@ ], "soft_pairs": [ "issue:44623|issue:45397", + "issue:43441|issue:44683", "issue:44623|issue:44749", - "issue:43262|issue:44521", "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:42175|issue:43976", "issue:44671|issue:44779", "issue:43784|issue:45542", "issue:43901|issue:43994", @@ -10491,232 +10388,218 @@ "issue:44188|issue:44336", "issue:43525|issue:43881", "issue:43525|issue:43643", - "issue:43295|issue:44617", + "issue:43450|issue:44617", "issue:44589|issue:45362", "issue:41628|issue:43976", "issue:45561|issue:45563", - "issue:42491|issue:43299", "issue:43901|issue:44246", + "issue:42491|issue:43299", "issue:44336|issue:44373", "issue:44242|issue:45593", - "issue:44295|issue:45397", - "issue:41950|issue:44220", - "issue:44246|issue:44496" + "issue:44295|issue:45397" ], - "prepared_review_unit_hash": "9b9c172da8e49c237f907793dc8afa752614c469a0ad7ebeb64a6f87c309f09b", + "prepared_review_unit_hash": "89f96f0701692593d902868c841374adc4472e841bb338ecc00da64aff6446f1", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36400, - "estimated_input_tokens": 9100, - "estimated_eval_tokens": 18456 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35460, + "estimated_input_tokens": 8865, + "estimated_eval_tokens": 17986 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of distinct Transformers v5 regressions and integration issues across different models, processors, tokenizers, and backends. I did not find any soft pair that clearly describes the same underlying bug or change, so the safe triage choice is to keep them separate.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43723 is the broadest and most representative report here: a generic AutoTokenizer.from_pretrained loading failure in v5, which fits the cluster\u2019s overall theme of v5 regressions without being too model-specific.", + "summary": "The cluster is mostly a loose embedding match around Transformers v5-era regressions, but the paired items are on different code paths (tokenizers, model loading, chat templates, attention kernels, backend packaging, docs, etc.). I do not see any soft pair that is close enough to be treated as the same underlying bug/change.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44521 is the most central user-facing bug in this set: it is broad, clearly described, open, and has the most obvious potential to attract/collect related multimodal chat-template regressions.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43723 is the best global representative because it is clear, central, and broadly phrased enough to anchor this cluster while still describing a concrete failure mode.", + "best_issue_reason": "issue:44521 is the best representative issue here because it is concrete, high-signal, and more likely than the others to serve as a meaningful anchor for nearby regressions.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Different problems: missing files from processor.save_pretrained vs gemma-4 zero3 from_pretrained loading failure." + "reason": "Different failures: saving processor artifacts vs loading a model under zero3; no shared concrete code-path." }, { - "left": "issue:44623", - "right": "issue:44749", + "left": "issue:43441", + "right": "issue:44683", "accept": false, - "reason": "One is a processor save_pretrained bug; the other is a performance regression during data filtering after upgrade." + "reason": "Both mention attention backends, but one is FlashAttention in Ministral-3 and the other is compiled flex_attention on torch>=2.9; distinct bugs." }, { - "left": "issue:43262", - "right": "issue:44521", + "left": "issue:44623", + "right": "issue:44749", "accept": false, - "reason": "Both involve apply_chat_template, but one is about audio sampling rate defaults and the other about zero assistant masks for multimodal inputs." + "reason": "Processor save_pretrained missing files and tokenizer throughput regression are unrelated." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "Unrelated model bugs: BigBirdTokenizer special-token registration vs D-FINE auxiliary losses when denoising is off." + "reason": "BigBirdTokenizer special-token registration is unrelated to D-FINE auxiliary loss logic." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Same API family, but different bug: audio sampling-rate default vs multimodal assistant mask generation." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Different failure modes and subsystems: generic tokenizer loading vs ESM2 model breakage." + "reason": "Tokenizer loading regression and ESM2 model breakage are different issues." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error is unrelated to Gemma4 device-map forward mismatch." + "reason": "Notebook syntax error is not the same as Gemma4 device-map forward mismatch." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache/chunked_prefill issue is unrelated to CLIPVisionModel hidden_states handling." + "reason": "Generation cache/chunked_prefill failure and CLIPVision hidden_states regression affect different modules and behaviors." }, { - "left": "issue:42175", - "right": "issue:43976", + "left": "issue:44361", + "right": "issue:45593", "accept": false, - "reason": "Backend packaging omission for TensorFlow under a torch install is not the same as the Python version compatibility bug." + "reason": "Tokenizer task AttributeError and D-FINE auxiliary loss omission are unrelated." }, { - "left": "issue:44361", - "right": "issue:45593", + "left": "issue:42175", + "right": "issue:43976", "accept": false, - "reason": "MLukeTokenizer task AttributeError and D-FINE auxiliary-loss behavior are different code paths." + "reason": "Backend packaging/install issue and Python version compatibility issue are different root causes." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "Both are v5 correctness regressions, but they affect different model/tokenizer families and do not look like one concrete bug." + "reason": "Different model families and different symptoms: CamemBERT MLM predictions vs Deepseek tokenization." }, { "left": "issue:43784", "right": "issue:45542", "accept": false, - "reason": "Importing sentence-transformers with nn missing is unrelated to tensorboard-only installation causing undefined tf backend errors." + "reason": "Both are import/backend-related, but one is a missing nn import in sentence-transformers integration and the other is a missing TensorFlow backend when only tensorboard is installed." }, { "left": "issue:43901", "right": "issue:43994", "accept": false, - "reason": "Docs mismatch for return_all_scores is not the same as incorrect SigLIP2 model/pipeline outputs." + "reason": "Docs mismatch about return_all_scores is not the same as incorrect SigLIP2 inference outputs." }, { "left": "issue:43122", "right": "issue:43650", "accept": false, - "reason": "Different scope entirely: tokenizer output changes vs a nonspecific data request." + "reason": "Tokenizer output drift and a nonspecific 'ADD THE DATA' issue do not identify the same bug." }, { "left": "issue:44987", "right": "issue:45542", "accept": false, - "reason": "Loading a specific model on transformers>=5.1.0 is unrelated to the tensorboard/TensorFlow backend packaging issue." + "reason": "Model loading failure for a specific repo and TensorFlow backend packaging issue are unrelated." }, { "left": "issue:43824", "right": "issue:43976", "accept": false, - "reason": "Missing Qwen2.5-VL import and Python 3.9/3.10 compatibility are unrelated failures." + "reason": "Qwen2_5_VL import failure and Python 3.9/3.10 compatibility are different failure modes." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer class metadata written by save_pretrained is unrelated to Gemma4 zero3 loading." + "reason": "Tokenizer config serialization mismatch and Gemma4 loading under zero3 are different code paths." }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "torch.compile attention kernel divergence and ANSI codes in loading_report are different subsystems." + "reason": "Attention-kernel divergence under torch.compile is unrelated to ANSI codes in loading_report." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Missing pad_token_id in Llama4Config is not the same as generic glm-4v-9b loading failure." + "reason": "Llama4Config missing pad_token_id and glm-4v-9b loading failure are model-specific and not the same bug." }, { "left": "issue:43525", "right": "issue:43643", "accept": false, - "reason": "Both mention missing fields, but one is a specific config attribute bug and the other is trust_remote_code stripping fields during AutoConfig loading." + "reason": "One is a specific config attribute absence; the other is a generic trust_remote_code field-loss issue. Similar theme, different defect." }, { - "left": "issue:43295", + "left": "issue:43450", "right": "issue:44617", "accept": false, - "reason": "Processor/tokenizer regression is unrelated to Sam3Video CUDA OOM." + "reason": "Incorrect batched video processor shapes and Sam3Video CUDA OOM are not the same underlying problem." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Float8 storage lookup error and Qwen3.5-35B chat crash are different runtime failures." + "reason": "Float8 storage lookup failure and Qwen3.5 chat crash are different runtime errors." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Missing AutoImageProcessor import is unrelated to the Python version compatibility issue." + "reason": "Importing AutoImageProcessor and Python version support are unrelated." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "pytest-xdist file-race bug and stale generation warning are unrelated." + "reason": "pytest-xdist file race and stale generate() warning are unrelated test/runtime issues." }, { - "left": "issue:42491", - "right": "issue:43299", + "left": "issue:43901", + "right": "issue:44246", "accept": false, - "reason": "Both mention Qwen3 MoE, but one is LoRA compatibility across branches and the other is base model loading for Qwen3VL; too different to merge confidently." + "reason": "Documentation note about return_all_scores and slow import behavior are different concerns." }, { - "left": "issue:43901", - "right": "issue:44246", + "left": "issue:42491", + "right": "issue:43299", "accept": false, - "reason": "Documentation mismatch for text classification is unrelated to a runtime performance report about import time." + "reason": "Both involve Qwen3 MoE, but one is LoRA compatibility across versions and the other is base-model loading regression; too different to merge as one fix." }, { "left": "issue:44336", "right": "issue:44373", "accept": false, - "reason": "ANSI escape output and a wrong docstring are not the same bug." + "reason": "ANSI output formatting and a wrong docstring are not the same bug." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Both concern losses, but they are for different models and different gating conditions." + "reason": "Load-balancing loss routing logic and D-FINE auxiliary-loss behavior are unrelated." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "position_ids buffer access error is unrelated to Gemma4 zero3 from_pretrained." - }, - { - "left": "issue:41950", - "right": "issue:44220", - "accept": false, - "reason": "Video-classification processor mismatch and fbank feature extraction error are different pipeline/audio issues." - }, - { - "left": "issue:44246", - "right": "issue:45593", - "accept": false, - "reason": "Import latency complaints are unrelated to D-FINE auxiliary-loss behavior." - }, - { - "left": "issue:43976", - "right": "issue:44496", - "accept": false, - "reason": "Python version support regression is unrelated to unrecognized model/config loading for Olmo-Hybrid-Instruct-SFT-7B." + "reason": "Buffer registration/readback error and Gemma4 zero3 loading failure are different defects." } ] }, @@ -10735,12 +10618,13 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ "issue:36246", + "issue:41950", "issue:42757", "issue:42886", "issue:43065", @@ -10749,17 +10633,15 @@ "issue:43262", "issue:43352", "issue:43408", - "issue:43475", "issue:43526", "issue:43531", - "issue:43540", "issue:43596", - "issue:43653", "issue:43673", "issue:43901", "issue:44060", "issue:44077", "issue:44162", + "issue:44220", "issue:44242", "issue:44246", "issue:44295", @@ -10790,6 +10672,8 @@ "issue:45759" ], "soft_pairs": [ + "issue:41950|issue:44220", + "issue:44246|issue:44496", "issue:43531|issue:45362", "issue:36246|issue:42757", "issue:43352|issue:45250", @@ -10817,218 +10701,216 @@ "issue:43257|issue:45137", "issue:43262|issue:44246", "issue:44995|issue:45200", - "issue:44060|issue:44077", - "issue:43475|issue:43540", - "issue:43653|issue:44162" + "issue:44060|issue:44077" ], - "prepared_review_unit_hash": "f3dfcd15ff27b96e20056596fcb9f9a463eb897d24e313b63df433d48a485e3d", + "prepared_review_unit_hash": "70abf0262e597b254f4f0fd35406be8df6929df57bd75646db8323e6ed6ffa78", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36604, - "estimated_input_tokens": 9151, - "estimated_eval_tokens": 18558 + "serialized_chars": 35958, + "estimated_input_tokens": 8990, + "estimated_eval_tokens": 18236 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a mixed bag of mostly unrelated Transformers issues, with a few loose thematic overlaps around imports, configs, and model loading. I did not find any soft pair that looks like the same concrete bug/change, so I rejected all candidate merges.", - "confidence": 0.86, - "canonical_issue_reason": "issue:44246 is the broadest, most central user-facing regression in the set and has the strongest discussion activity; it best represents the cluster\u2019s import/loading/core-utility theme better than the model-specific reports.", + "summary": "No soft pair looks like the same underlying bug or change. The set is heterogeneous across unrelated areas (pipelines, tokenizer/config loading, generation, DeepSpeed, and model-specific issues), so none should be deduplicated.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue fits this cluster: the candidates cover unrelated defects and no pair looks mergeable as the same bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44246 is the best overall issue candidate because it is broad, high-activity, and closest to a cluster-level core library problem.", + "best_issue_reason": "No global best issue for deduplication; the list is too mixed to anchor a canonical representative.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Different subsystems and symptoms: video-classification pipeline processor lookup vs audio fbank feature extraction." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import performance issue is unrelated to model config parsing failure for an unrecognized model." + }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Both involve Qwen models, but one is a sliding_window bug and the other is a chat/runtime crash; different code paths and symptoms." + "reason": "Both involve Qwen, but the bugs differ: sliding-window handling vs chat crash on Qwen3.5-35B." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but they refer to different missing symbols in different packages; not the same underlying bug." + "reason": "Distinct import errors in different packages: Qwen2.5-VL image processor vs huggingface_hub offline helper." }, { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "One is Nemotron-specific Flash Attention 2.0 support, the other is a generic Flash Attention topic; too broad to be the same fix." + "reason": "One is a model-specific Flash Attention support error; the other is only a broad feature title with no matching code path shown." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Different missing imports (`Qwen2_5_VLImageProcessor` vs `HybridCache`) and different subsystems." + "reason": "Different missing imports from different modules; no shared code-path or fix." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "Completely different problems: terminal ANSI output vs GLM-5 RoPE implementation." + "reason": "ANSI color output in loading reports is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "One is a BeitImageProcessorFast label-reduction bug; the other is unsafe `sys.modules` access in modeling_utils." + "reason": "BEiT image processor label reduction bug is unrelated to unsafe sys.modules access in modeling_utils." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Core utility import/access bug versus a pytest-xdist file race; unrelated failure modes." + "reason": "Unsafe module lookup and xdist file race are separate testing/runtime problems." }, { "left": "issue:43673", "right": "issue:44561", "accept": false, - "reason": "Both mention v5 regressions, but one is generation cache/chunked prefill and the other is `is_torch_fx_available` removal breaking trust_remote_code models." + "reason": "Chunked-prefill cache bug is unrelated to removal of is_torch_fx_available breaking trust_remote_code models." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import latency and Gemma eos_token_id mismatch are unrelated." + "reason": "Import slowdown has no concrete relation to Gemma eos_token_id configuration mismatch." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "Different model losses and different failure modes; no shared code path." + "reason": "Load-balancing loss omission and LWDETR AMP crash are different MoE vs vision-loss bugs." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage and Gemma4 token-type defaults are unrelated model-specific issues." + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaulting are unrelated model-specific issues." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "Registering `position_ids` as a buffer and a slowdown after upgrade are not the same bug." + "reason": "Position_ids buffer access error is unrelated to the reported tokenization slowdown." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "ImportError for a processor class versus a ZeRO3/BertModel IndexError; different layers of the stack." + "reason": "Qwen2.5-VL import failure and DeepSpeed ZeRO3 deque IndexError are unrelated." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Both concern strict config handling, but they affect different model families and different missing fields." + "reason": "Strict config loading for granite_speech is not the same as missing Qwen3_5MoeVisionConfig field handling." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Tokenizer behavior regression versus resize_token_embeddings not updating output embeddings; not the same concrete issue." + "reason": "Tokenizer regression and resize_token_embeddings output embedding mismatch are different bugs." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Output embedding resize bug and Whisper empty-transcription generation failure are unrelated." + "reason": "Embedding resize behavior is unrelated to Whisper empty-transcription generation failure." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Documentation mismatch versus low-level `sys.modules` access bug." + "reason": "Docs mismatch for return_all_scores is unrelated to sys.modules access." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "Resize-token-embeddings behavior is not the same as a general tokenization-version-change report." + "reason": "resize_token_embeddings/output embeddings bug is not the same as version-dependent tokenization changes." }, { "left": "issue:43232", "right": "issue:44336", "accept": false, - "reason": "Generation kwargs/update logic and ANSI-reporting output are unrelated." + "reason": "Generation kwargs update after sync_gpus is unrelated to ANSI codes in loading_report." }, { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Offline cache loading and Gemma4 multi-GPU OOM are different problems with no shared code path." + "reason": "Offline tokenizer cache loading bug is unrelated to Gemma4 multi-GPU OOM." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "Both are SAM3-related, but one is a dummy Conv2d in the pixel decoder and the other is a model-type/config mismatch warning." + "reason": "Dummy Conv2d in Sam3PixelDecoder and sam3_video/sam3_tracker config mismatch are different SAM3 issues." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Loading a specific model on v5.1.0 and protobuf decode-error handling are unrelated." + "reason": "Model loading failure for physical-intelligence/fast is unrelated to tokenizer exception masking when protobuf is absent." }, { "left": "issue:44496", "right": "issue:45003", "accept": false, - "reason": "Unrecognized model/config loading and unsafe `sys.modules` access are distinct issues." + "reason": "Unrecognized model/config error is unrelated to unsafe sys.modules access." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Both are Qwen composite-config issues, but one drops a specific vision config field while the other fails to unwrap `text_config`; not the same bug." + "reason": "Missing config field in Qwen3_5MoeVisionConfig is not the same as AutoModelForCausalLM unwrapping text_config." }, { "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 underflow; different failure paths." + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed and ZeRO3 deque popping are different failure modes." }, { "left": "issue:43262", "right": "issue:44246", "accept": false, - "reason": "Audio processor sampling-rate defaults and import performance are unrelated." + "reason": "Audio chat-template sample-rate default bug is unrelated to import performance." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "A stale indexer cache on second forward pass is not the same as Gemma4 token-type defaults." + "reason": "GlmMoeDsa stale indexer cache crash does not match Gemma4 token-type-id defaults." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Different models and different issues: tied-weights warning versus optional post_init handling." - }, - { - "left": "issue:43475", - "right": "issue:43540", - "accept": false, - "reason": "Both are multimodal/video-adjacent, but one is a missing vision-encoder attribute and the other is a video-input processing ValueError." - }, - { - "left": "issue:43653", - "right": "issue:44162", - "accept": false, - "reason": "BigBirdTokenizer special-token registration bug and ESM2 breakage are unrelated." + "reason": "Qwen3.5 tied-weights warning and patchtsmixer optional post_init are unrelated model/config issues." } ] }, @@ -11047,8 +10929,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -11056,16 +10938,16 @@ "issue:30990", "issue:41628", "issue:42371", - "issue:42673", "issue:42994", "issue:43232", "issue:43475", "issue:43482", "issue:43525", + "issue:43540", "issue:43643", "issue:43646", + "issue:43653", "issue:43673", - "issue:43704", "issue:43931", "issue:44038", "issue:44162", @@ -11102,11 +10984,13 @@ "issue:45656" ], "soft_pairs": [ + "issue:43475|issue:43540", + "issue:43653|issue:44162", "issue:44466|issue:45335", "issue:45397|issue:45656", + "issue:44995|issue:45588", "issue:44368|issue:45563", "issue:44368|issue:44977", - "issue:44995|issue:45588", "issue:43673|issue:44734", "issue:45137|issue:45600", "issue:43931|issue:44368", @@ -11122,8 +11006,8 @@ "issue:44749|issue:45310", "issue:28282|issue:30990", "issue:42994|issue:43482", - "issue:42371|issue:45092", "issue:45230|issue:45397", + "issue:42371|issue:45092", "issue:43475|issue:43643", "issue:43475|issue:45381", "issue:44038|issue:44623", @@ -11132,246 +11016,241 @@ "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:43232|issue:44188" ], - "prepared_review_unit_hash": "0c74d2e61043e038ad2dd9837dc7aec042179d71516baf5ff44d86548c574127", + "prepared_review_unit_hash": "589c4b661defa5488cb3fea9cb22746fb1831c6b1d232049676fe27c1890884f", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37570, - "estimated_input_tokens": 9393, - "estimated_eval_tokens": 19042 + "serialized_chars": 37588, + "estimated_input_tokens": 9397, + "estimated_eval_tokens": 19050 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous v5 regression reports and not one single duplicate set. The only clear duplicate groups are the repeated Qwen3ForCausalLM VRAM leak report and the Qwen3.5-Moe from_pretrained/config-field failures.", - "confidence": 0.69, - "canonical_issue_reason": "45278 is the broadest, most umbrella-style report in the set: a general import/regression breakage after upgrading to 5.5.0. It is the best single anchor for the cluster because many other items are narrower one-off model or subsystem failures.", + "summary": "The set is mostly a grab bag of unrelated regression reports across tokenizers, generation, model loading, deepspeed, vision, and import/export paths. I did not find any soft pair that looks like the same concrete bug fix target, so no duplicate merges are warranted.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45278 is the broadest current umbrella report in the set: it targets upgrade-related import regressions across many cases and is open, making it the best central triage anchor.", "canonical_pr_reason": null, - "best_issue_reason": "45278 is the most useful issue to triage first because it captures the widest shared symptom pattern and is open, while most other items are narrower, model-specific, or unrelated to the main regression theme.", + "best_issue_reason": "issue:45278 is the best single issue to keep as the representative hub because it is broad, current, and likely to accumulate related import-regression reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "Different models and failure modes: SAM 3 video missing an encoder output attribute vs Qwen3OmniMoe video-input processing ValueError." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "One is a BigBirdTokenizer special-token registration bug; the other is an ESM2 breakage report affecting workflows, not the same code path." + }, { "left": "issue:44466", "right": "issue:45335", "accept": false, - "reason": "Both touch tied-embedding/serialization behavior, but one is about lm_head weight serialization and the other about resize_token_embeddings not propagating to decoder.embed_tokens. Related area, different bug." + "reason": "Both involve weight/embedding serialization, but one is tied lm_head serialization by device and the other is resize_token_embeddings not updating decoder.embed_tokens; different bugs." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "One is a Gemma-4 from_pretrained/Zero3 loading failure; the other is an optimizer step being invoked twice under DeepSpeed. Different code paths and symptoms." + "reason": "Both mention DeepSpeed, but one is a gemma-4 from_pretrained/ZeRO3 issue and the other is double optimizer stepping; unrelated code paths." }, { - "left": "issue:44368", - "right": "issue:45563", + "left": "issue:44995", + "right": "issue:45588", "accept": false, - "reason": "Both mention warnings, but they concern unrelated behaviors: tied-embedding warnings during Qwen fine-tuning versus a stale num_return_sequences warning in paged generation." + "reason": "A second-forward stale indexer cache crash is not the same as a flash-attention crash on s_aux=None." }, { "left": "issue:44368", - "right": "issue:44977", + "right": "issue:45563", "accept": false, - "reason": "The first is about a tie_word_embeddings warning; the second is a Qwen3.5 flash-attention generation failure. Same model family is not enough to merge them." + "reason": "One is a tie_word_embeddings warning during fine-tuning; the other is a stale num_return_sequences warning in paged generate()." }, { - "left": "issue:44995", - "right": "issue:45588", + "left": "issue:44368", + "right": "issue:44977", "accept": false, - "reason": "A stale indexer cache crash in GlmMoeDsa is unrelated to a flash_attention AttributeError on s_aux=None." + "reason": "A warning about tie_word_embeddings is unrelated to Qwen3.5 generation failures under flash-attention." }, { "left": "issue:43673", "right": "issue:44734", "accept": false, - "reason": "Both are generation/cache bugs, but one is a missing GenerationMixin cache in chunked_prefill and the other is wrong tensor indexing in serve KV-cache continuation. Different failures." + "reason": "Both touch KV/cache behavior, but one is missing GenerationMixin cache in chunked_prefill and the other is wrong tensor indexing during /v1/responses continuation." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "DeepSpeed ZeRO3 deque underflow and a broken auto_mappings reference to removed Sam3LiteText configs are unrelated." + "reason": "DeepSpeed ZeRO3 deque underflow is unrelated to stale auto_mappings references to removed Sam3LiteText configs." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and a tie_word_embeddings warning in Qwen3.5 fine-tuning are not the same bug." + "reason": "Qwen3-VL weight-shape mismatch is unrelated to a config warning about tie_word_embeddings." }, { "left": "issue:44933", "right": "issue:45459", "accept": false, - "reason": "Both are import/tokenizer-related, but one is a missing image_utils import and the other is protobuf error handling hiding tokenizer parse errors. Different fixes." + "reason": "A missing import from image_utils is not the same as protobuf decode errors being masked by except import_protobuf_decode_error()." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper trailing-replacement decoding crash and Kimi-K2.5 codec/fix_mistral_regex regression are distinct tokenizer issues." + "reason": "Whisper trailing-replacement-character timestamp decoding and Kimi-K2.5 codec/fix_mistral_regex regression are different tokenizer issues." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "Both are generation internals, but they affect different states: sync_gpus/update_model_kwargs versus chunked_prefill cache availability." + "reason": "Both are generation-related, but one is sync_gpus/update_model_kwargs logic and the other is v5 chunked_prefill cache availability." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "tokenizer.save_pretrained writing the wrong tokenizer_class and processor.save_pretrained missing files are separate serialization bugs." + "reason": "tokenizer.save_pretrained writing the wrong tokenizer_class is a different save-path bug from processor.save_pretrained missing files." }, { "left": "issue:43673", "right": "issue:45278", "accept": false, - "reason": "A generation cache bug is not the same as broad import failures after upgrading to v5.5.0." + "reason": "Chunked_prefill cache handling is not the same as broad import errors after upgrading from 4.57.0 to 5.5.0." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "AutoConfig remote-code missing fields and an ESM2 broken-workflow report do not share the same underlying defect." + "reason": "Custom model initialization breakage in v5 is too broad to equate with an ESM2-specific workflow regression." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer task AttributeError and a Chinese report about slower filtering after upgrade are unrelated." + "reason": "MLukeTokenizer AttributeError on tasks is unrelated to performance slowdown in filtering after upgrading Transformers." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Missing processor save files and a DeepSpeed double-step optimizer bug are different subsystems and symptoms." + "reason": "Missing files from processor.save_pretrained is unrelated to DeepSpeed causing the optimizer step to run twice." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Performance regression after upgrade and DeepSpeed optimizer double-step are not the same issue." + "reason": "A performance regression in data filtering and a DeepSpeed double-step bug are not the same underlying issue." }, { "left": "issue:44749", "right": "issue:45310", "accept": false, - "reason": "The filtering slowdown report and Qwen3.5-Moe from_pretrained failure are unrelated." + "reason": "Slow filtering after upgrade is unrelated to Qwen3.5 Moe from_pretrained errors." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "AutoModel PyTorch ImportError and Sentence Transformers stuck loading are both environment/loading complaints, but not the same underlying bug." + "reason": "PyTorch-missing ImportError and Sentence Transformers hanging on load are different environment/loading problems." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different quantization/load-path problems." + "reason": "Quantized model saving failure is not the same as Qwen2.5-GGUF loading failure in v5." }, { - "left": "issue:42371", - "right": "issue:45092", + "left": "issue:45230", + "right": "issue:45397", "accept": false, - "reason": "TF32 API settings guidance and old InternVL2 remote-code meta-init incompatibility are unrelated." + "reason": "A placeholder bug report does not identify the same from_pretrained/ZeRO3 issue as gemma-4." }, { - "left": "issue:45230", - "right": "issue:45397", + "left": "issue:42371", + "right": "issue:45092", "accept": false, - "reason": "A placeholder bug report and a specific Gemma-4 loading issue are not duplicates." + "reason": "TF32 API-setting guidance is unrelated to old InternVL2 checkpoints breaking under meta initialization." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "SAM 3 Video missing attribute and AutoConfig trust_remote_code field loss are different model/config problems." + "reason": "SAM 3 video encoder output missing an attribute is unrelated to AutoConfig returning missing fields under trust_remote_code=True." }, { "left": "issue:43475", "right": "issue:45381", "accept": false, - "reason": "SAM 3 Video missing fpn_position_embeddings and Qwen2.5-VL video position-id errors are unrelated model-specific bugs." + "reason": "Different vision/video regressions in different models: SAM 3 output attribute error vs Qwen2.5-VL video position IDs." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe v5 bug and processor.save_pretrained missing files do not share the same failure mode." + "reason": "Qwen3-VL-Moe from_pretrained breakage and processor.save_pretrained missing files are separate bugs." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "Both are import errors, but one is AutoImageProcessor export/import and the other is a missing import from image_utils; not the same symbol or root cause." + "reason": "An AutoImageProcessor import error and a missing image_utils import are related only at a broad import level, not the same concrete bug." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Mistral regex patching crash and docstring-annotation AttributeError are unrelated." + "reason": "Mistral regex patch crashing on tokenizer backend access is unrelated to auto_docstring crashing on future annotations." }, { "left": "issue:43475", "right": "issue:44661", "accept": false, - "reason": "SAM 3 Video attribute error and add-new-model-like failure inside TOKENIZER_MAPPING_NAMES are different issues." + "reason": "SAM 3 vision encoder output and add-new-model-like failing inside TOKENIZER_MAPPING_NAMES are different subsystems and failures." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized save failure and Llama4Config missing pad_token_id are separate bugs." + "reason": "Quantized model saving and Llama4Config missing pad_token_id are unrelated." }, { "left": "issue:45375", "right": "issue:45478", - "accept": true, - "reason": "Both are Qwen3.5-Moe from_pretrained/config-loading failures around the same v5-era regression; the missing deepstack_visual_indexes field is a plausible root cause of the reported loading error." + "accept": false, + "reason": "Missing deepstack_visual_indexes in a vision config is not the same as Qwen3.5 Moe from_pretrained errors." }, { "left": "issue:43232", "right": "issue:44188", "accept": false, - "reason": "GenerationMixin cache handling after sync_gpus and divergent attention kernels under torch.compile are different problems." - }, - { - "left": "issue:42673", - "right": "issue:43704", - "accept": true, - "reason": "These have the same title and describe the same Qwen3ForCausalLM VRAM leak when used across multiple dataloader threads." - }, - { - "left": "issue:45310", - "right": "issue:45375", - "accept": true, - "reason": "Both point to the same Qwen3.5-Moe from_pretrained regression; one is the generic loading error and the other is the missing config field that likely triggers it." + "reason": "GenerationMixin cache handling after sync_gpus is unrelated to diverging attention kernels under torch.compile." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The duplicate calls are mostly well-supported by title/body alignment, especially the exact Qwen3ForCausalLM VRAM-leak pair and the closely related Qwen3.5-Moe from_pretrained/config-field reports. The umbrella choice of issue 45278 is also reasonable given the broad import-regression wording and open state." - }, - "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 }, @@ -11388,14 +11267,13 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ - "issue:30333", "issue:30990", - "issue:33290", "issue:35141", "issue:36010", "issue:41628", "issue:42175", "issue:42222", + "issue:42673", "issue:42831", "issue:42898", "issue:42994", @@ -11406,7 +11284,7 @@ "issue:43504", "issue:43526", "issue:43673", - "issue:43723", + "issue:43704", "issue:43784", "issue:43824", "issue:43873", @@ -11432,16 +11310,19 @@ "issue:45137", "issue:45200", "issue:45310", + "issue:45375", "issue:45561", "issue:45712", "issue:45753", "issue:45812" ], "soft_pairs": [ + "issue:42673|issue:43704", + "issue:45310|issue:45375", "issue:43873|issue:44945", "issue:41628|issue:44351", - "issue:44945|issue:45753", "issue:42831|issue:44222", + "issue:44945|issue:45753", "issue:44222|issue:45310", "issue:43673|issue:44188", "issue:36010|issue:41628", @@ -11466,231 +11347,232 @@ "issue:44623|issue:45310", "issue:42898|issue:45812", "issue:43278|issue:45137", - "issue:44734|issue:45103", - "issue:43723|issue:44568", - "issue:30333|issue:33290" + "issue:44734|issue:45103" ], - "prepared_review_unit_hash": "4dc3c803cdfd134d65f5acce3a9e97efe46c8baa9a54c6dd49d90938885af8d0", + "prepared_review_unit_hash": "730eb66fc0268471083c5f17f988d6e7d699dda0f25f9a5955360f320d9e2a11", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36764, - "estimated_input_tokens": 9191, - "estimated_eval_tokens": 18638 + "serialized_chars": 36787, + "estimated_input_tokens": 9197, + "estimated_eval_tokens": 18650 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous and does not contain any strong duplicate pairs. Most items are unrelated transformer regressions, but the broadest representative is the v5 tokenizer-loading regression. No PRs are present.", - "confidence": 0.69, - "canonical_issue_reason": "Issue 43723 is the broadest and most generic of the set, covering a v5 tokenizer-loading regression that is adjacent to several other tokenizer-related reports. The rest of the cluster is too diverse for a single true canonical issue.", + "summary": "This cluster is heavily over-broadened: most items are unrelated issues that only share superficial subsystem or model-family similarity. The only clear duplicate pair is the Qwen3ForCausalLM VRAM leak report.", + "confidence": 0.88, + "canonical_issue_reason": "issue:42673 is the clearest representative of the only true duplicate pair, with a detailed report and higher discussion activity than its twin.", "canonical_pr_reason": null, - "best_issue_reason": "43723 best represents the cluster only as a loose umbrella for v5 tokenizer regressions; however, it does not truly subsume the many unrelated import, multimodal, FP8, and save/load issues here.", + "best_issue_reason": "issue:42673 is the strongest global issue candidate because it matches an exact duplicate, has the fuller description, and appears earlier than the matching report.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Exact same title and same underlying bug: Qwen3ForCausalLM VRAM leak in multiple dataloader threads." + }, + { + "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; too different to merge." + }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Both involve bad behavior in model execution, but one is quantization/offloading and the other is pipeline-parallel output correctness; different code paths and symptoms." + "reason": "Different failures: quantization offloading vs incorrect outputs under pipeline parallelism." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they are for different missing symbols (`AutoImageProcessor` vs `HybridCache`) and likely different export gaps." + "reason": "Both are import errors, but for different symbols and different code paths." }, { - "left": "issue:44945", - "right": "issue:45753", + "left": "issue:42831", + "right": "issue:44222", "accept": false, - "reason": "Pipeline-parallel incorrect output is not the same bug as a Qwen3_5 image-specific infinite loop; different failure modes and likely different fixes." + "reason": "Accuracy regression in FineGrainedFP8 is not the same bug as FP8 save_pretrained failure." }, { - "left": "issue:42831", - "right": "issue:44222", + "left": "issue:44945", + "right": "issue:45753", "accept": false, - "reason": "One is an FP8 accuracy regression, the other is a save_pretrained failure for MoE FP8; related area but not the same underlying bug." + "reason": "Pipeline-parallel output corruption and Qwen3_5 image infinite loop are unrelated symptoms." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen3.5/MoE and loading/saving, but one is FP8 save_pretrained and the other is a from_pretrained error in newer transformers; not the same concrete issue." + "reason": "Saving a quantized MoE model and loading Qwen3.5 MoE with from_pretrained are different concrete problems." }, { "left": "issue:43673", "right": "issue:44188", "accept": false, - "reason": "Generation cache missing in chunked prefill and attention-kernel divergence under torch.compile are separate generation/attention bugs." + "reason": "Generation cache handling during chunked_prefill is unrelated to attention-kernel divergence under torch.compile." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "Both are import failures, but they concern different APIs and likely unrelated compatibility regressions." + "reason": "Different missing imports: GenerationMixin versus AutoImageProcessor." }, { "left": "issue:42994", "right": "issue:44164", "accept": false, - "reason": "Both involve serialization, but one is quantized model saving and the other is extra_state handling in save/from_pretrained; not the same bug." + "reason": "Quantized model saving and extra_state save/from_pretrained handling are different serialization bugs." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "One is packaging/backends missing TensorFlow under `.[torch]`, the other is leftover dummy class cleanup; different concerns." + "reason": "Packaging/backend dependency issue is not the same as leftover dummy classes leaking into dir(transformers)." }, { "left": "issue:42222", "right": "issue:45200", "accept": false, - "reason": "VitPose model breakage and Gemma 4 mm_token_type_ids defaulting are unrelated model-specific bugs." + "reason": "VitPose breakage and Gemma 4 mm_token_type_ids behavior are unrelated model-specific issues." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Both touch generation, but one is a sync_gpus bookkeeping issue and the other is a Janus image-generation test failure; not the same code-path problem." + "reason": "Generation kwargs update after sync_gpus is unrelated to a janus image-generation test failure." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen3.5-VL/MoE, but one is a general transformers 5.0 bug report and the other is a specific from_pretrained regression in 5.4.0+." + "reason": "Different model families and different load failures; too broad to treat as one bug." }, { "left": "issue:43377", "right": "issue:44242", "accept": false, - "reason": "MIMI padding-mask mismatch and load-balancing loss omission are different model logic bugs." + "reason": "Missing padding-mask support in MIMI encoder is unrelated to MoE load-balancing loss computation." }, { "left": "issue:43526", "right": "issue:45561", "accept": false, - "reason": "Beit label reduction bug and xdist file-race in testing utils are unrelated." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to the xdist captured_info.txt race." }, { "left": "issue:43504", "right": "issue:45561", "accept": false, - "reason": "Pretrained preset loading with a legacy field is unrelated to a pytest-xdist captured-info race." + "reason": "Legacy-field pretrained loading failure is unrelated to the testing-utils race." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Unrecognized model/config loading and a testing-utils file race are unrelated." + "reason": "Unrecognized model/config loading error is unrelated to the xdist file race." }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Phi-4 multimodal model loading and ZeRO3 deque underflow are different runtime failures." + "reason": "Phi-4 multimodal loading and DeepSpeed ZeRO3 deque error are unrelated." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "TensorFlow backend packaging and missing image_utils import are unrelated." + "reason": "Backend dependency installation and nonexistent image_utils import are unrelated." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Tokenization-space cleanup behavior and BOS/EOS special-token insertion are both tokenizer-related, but they are distinct behaviors and fixes." + "reason": "Both are tokenization regressions, but they affect different behaviors and models; not the same bug." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Sentence-transformers loading hang and embedding resize/tie-word-embedding reinit are unrelated." + "reason": "Sentence Transformers loading hang and embedding reinitialization in post_init are unrelated." }, { "left": "issue:43479", "right": "issue:44869", "accept": false, - "reason": "Multimodal config defaults and Whisper timestamp decoding crash are separate subsystems." + "reason": "Multimodal config defaults and Whisper timestamp decode crash are unrelated." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version compatibility and missing image_utils import are unrelated." + "reason": "Python version compatibility and missing import from image_utils are unrelated." }, { "left": "issue:43479", "right": "issue:44079", "accept": false, - "reason": "Config default initialization and ModelOutput key assignment are different core library behaviors." + "reason": "Config initialization bug and ModelOutput key assignment bug are unrelated." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Attention-kernel branching under compile and a wrong docstring for position_ids are not the same issue." + "reason": "Attention-kernel branching under torch.compile is unrelated to a docstring issue." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but one is a sentence-transformers `nn` NameError and the other is a missing Qwen2.5-VL export; different fixes." + "reason": "Missing nn import in sentence-transformers and missing Qwen2_5_VLForConditionalGeneration import are different import problems." }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "Processor save_pretrained missing files and Qwen3.5 from_pretrained error are unrelated loading/saving problems." + "reason": "processor.save_pretrained missing files is unrelated to Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Both concern tokenizer behavior in v5, but one is cleanup-space handling and the other is Granite token IDs; different concrete regressions." + "reason": "Tokenization behavior change and Granite token-ID regression are different concrete regressions." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Embedding dtype mismatch during evaluate and a ZeRO3 deque error are unrelated." + "reason": "Embedding dtype drift in evaluation is unrelated to the ZeRO3 deque error." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "Serving KV-cache indexing crash and auto-docstring annotation handling are separate code paths." - }, - { - "left": "issue:43723", - "right": "issue:44568", - "accept": false, - "reason": "Both are tokenizer regressions, but one is tokenizer loading and the other is add_special_tokens BOS/EOS behavior; not the same underlying bug." - }, - { - "left": "issue:30333", - "right": "issue:33290", - "accept": false, - "reason": "MLFlow reporting on failed training jobs and DeepSpeed/Adafactor OOM are unrelated." + "reason": "KV-cache continuation crash in transformers serve is unrelated to the auto_docstring annotation crash." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative. The one accepted pair, issue:42673 vs issue:43704, is supported by identical titles plus the fuller/earlier report on 42673. The remaining soft-edge rejections stay cautious and do not overclaim duplication from superficial similarity." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -11702,14 +11584,15 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ "issue:22355", "issue:28282", "issue:30333", + "issue:33290", "issue:34689", "issue:35141", "issue:36331", @@ -11738,12 +11621,11 @@ "issue:44484", "issue:44485", "issue:44514", + "issue:44568", "issue:44623", "issue:44779", - "issue:44871", "issue:44933", "issue:44977", - "issue:44991", "issue:45092", "issue:45103", "issue:45125", @@ -11757,6 +11639,8 @@ "issue:45834" ], "soft_pairs": [ + "issue:30333|issue:33290", + "issue:43723|issue:44568", "issue:43874|issue:45092", "issue:42898|issue:45356", "issue:44060|issue:45125", @@ -11776,9 +11660,9 @@ "issue:22355|issue:28282", "issue:30333|issue:36331", "issue:44297|issue:45310", + "issue:30333|issue:34689", "issue:43232|issue:45834", "issue:44222|issue:44484", - "issue:30333|issue:34689", "issue:43723|issue:44977", "issue:28282|issue:30333", "issue:44263|issue:45092", @@ -11789,236 +11673,246 @@ "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:45092|issue:45357" ], - "prepared_review_unit_hash": "722274dfd80e165ba9ff3ed0c03d704a868d85f1d6a6037c060294336ac05e02", + "prepared_review_unit_hash": "bf533617fb1ce050b8172560b7b6a38b804bd6acfb10c63eff1c7e7c236f00d6", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38123, - "estimated_input_tokens": 9531, - "estimated_eval_tokens": 19318 + "serialized_chars": 38105, + "estimated_input_tokens": 9527, + "estimated_eval_tokens": 19310 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: many issues are v5 regressions or import/load errors, but they point to different symbols, models, and failure modes. I don\u2019t see any soft pair that is clearly the same underlying bug, so I reject all soft edges. The most representative issue is the core tokenizer-behavior regression in v5.", - "confidence": 0.82, - "canonical_issue_reason": "issue:42898 is the cleanest, most general core-library regression here: it describes a concrete v5 behavior change in tokenizer cleanup, which is broader and more central than the model-specific or environment-specific reports.", + "summary": "These issues are only loosely related by broad Transformers themes (serialization/save_pretrained, tokenizer regressions, multimodal/model-loading bugs). None of the soft pairs look like the same underlying bug or change, so I rejected all of them. No PRs are present.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 44222 is the strongest concrete bug report in the loose save_pretrained/serialization subcluster, with the clearest actionable failure mode among the candidates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:42898 is the strongest standalone issue in the set: clearly scoped, reproducible, and broadly relevant to Transformers behavior rather than one model or example script.", + "best_issue_reason": "Issue 44222 is the best overall anchor here: it is a specific, bug-shaped report rather than a question or broad symptom, even though the surrounding items are not true duplicates.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow failure reporting is unrelated to the deepspeed/Adafactor OOM." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Different tokenizer regressions: load failure vs add_special_tokens/BOS-EOS behavior." + }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "Different problems: a missing image processor method vs. remote-code/meta-init incompatibility for a different model." + "reason": "Missing image-patch API and remote-code meta-init incompatibility are different bugs." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer regressions, but they describe different concrete failures and likely different fixes." + "reason": "v5 tokenization-space behavior change and Kimi codec handling/warning are distinct tokenizer issues." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Different Qwen issues: tied-weights warning vs. missing tensor-parallel plan." + "reason": "Incorrect tied-weights warning is not the same as a missing _tp_plan for tensor parallelism." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both involve tokenizers, but the reported behavior and likely root cause are different." + "reason": "Both involve tokenizers, but one is a v5 spacing change and the other is DeepSeek output corruption." }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "Unrelated: dependency version bump vs. leftover dummy class cleanup." + "reason": "A released-version dependency bump and leftover dummy classes are unrelated maintenance problems." }, { "left": "issue:42898", "right": "issue:44448", "accept": false, - "reason": "Both are v5 regressions, but one is tokenization cleanup and the other is a Pegasus output-change regression." + "reason": "Both are version regressions, but they affect different models and symptoms." }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Different code paths: buffer registration/read issue vs. optimizer stepping twice under deepspeed." + "reason": "Buffer-reading error and double optimizer-step behavior are unrelated code paths." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Different import/packaging issues affecting different modules and symptoms." + "reason": "Missing image_utils import and dummy_pt_objects leftovers are different import/export issues." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Both touch save_pretrained, but one is FP8 MoE serialization and the other is tokenizer config metadata mismatch." + "reason": "Both mention save_pretrained, but one is FP8 MoE saving and the other is tokenizer metadata mismatch." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer serialization metadata bug is unrelated to deepspeed optimizer stepping." + "reason": "Tokenizer save metadata and duplicate optimizer steps are unrelated." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Different example scripts and failures: multi-label classification output vs. local dataset loading." + "reason": "Example-script multi-label inference failure and local dataset loading failure are different problems." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Config initialization bug vs. non-persistent buffer initialization bug; not the same defect." + "reason": "Config initialization with None and junk-filled non-persistent buffers are different bugs." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both are save_pretrained-related, but one is MoE FP8 saving and the other is missing processor files." + "reason": "Both are serialization-related, but model save_pretrained FP8 behavior is not the same as processor.save_pretrained missing files." }, { "left": "issue:44206", "right": "issue:45103", "accept": false, - "reason": "Different subsystems entirely: feature extractor argument handling vs. auto-docstring annotation parsing." + "reason": "Unsupported feature argument in a feature extractor is unrelated to auto_docstring annotation crashes." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "Accuracy regression in FineGrainedFP8 is unrelated to processor.save_pretrained missing files." + "reason": "FineGrainedFP8 accuracy problems are unrelated to missing processor save files." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but they concern different symbols and packages." + "reason": "Different missing imports from different packages/modules." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Different missing-import causes: transformers.onnx module vs. missing PyTorch dependency." + "reason": "transformers.onnx import failure and missing PyTorch are separate import-environment issues." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "Training-reporting/MLflow issue is unrelated to CustomTrainer.compute_loss signature change." + "reason": "MLflow not reported on failure is unrelated to the CustomTrainer compute_loss signature breakage." }, { "left": "issue:44297", "right": "issue:45310", "accept": false, - "reason": "Tokenizer save metadata bug is unrelated to Qwen3.5 from_pretrained loading failure." + "reason": "Tokenizer save metadata mismatch and Qwen3.5 from_pretrained failure are different load/save bugs." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "Training-job reporting and Llama 3.2 Vision model loading regression are unrelated." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation kwargs propagation bug and Kosmos2.5 OCR index error are unrelated." + "reason": "Generation kwargs update after sync_gpus is not the same as an OCR long-input index error." }, { "left": "issue:44222", "right": "issue:44484", "accept": false, - "reason": "MoE FP8 save_pretrained bug is unrelated to the max_shard_size question." - }, - { - "left": "issue:30333", - "right": "issue:34689", - "accept": false, - "reason": "MLFlow reporting on failed training is unrelated to Llama 3.2 Vision model loading regression." + "reason": "A concrete save_pretrained bug and a question about max_shard_size are not duplicates." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Both are tokenizer/generation complaints, but one is AutoTokenizer loading and the other is flash-attention generation." + "reason": "Tokenizer loading in v5 and Qwen3.5 flash-attention generation failure are different issues." }, { "left": "issue:28282", "right": "issue:30333", "accept": false, - "reason": "Missing PyTorch import error is unrelated to MLFlow reporting on failed training jobs." + "reason": "Missing PyTorch during AutoModel import and MLflow failure reporting are unrelated." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "Different bugs: DSA indexer torch.split behavior vs. remote-code/meta-init checkpoint incompatibility." + "reason": "torch.split return handling and InternVL2 meta-init incompatibility are different model bugs." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "MLflow reporting on failure is unrelated to embedding reinitialization in post_init." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat-completions request rejection and Gemma4Processor missing _tokenizer are different API/processor failures." }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "Same broad DSA area, but torch.split return values and missing ReLU are distinct implementation issues." + "reason": "Same DSA indexer area, but one is a split-return bug and the other is a missing ReLU concern." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Docstring typo/incorrect docs is not the same as a RoPE implementation bug." + "reason": "A wrong docstring and a RoPE implementation discussion are not the same bug." }, { "left": "issue:44295", "right": "issue:44623", "accept": false, - "reason": "Buffer access error is unrelated to missing processor files in save_pretrained." + "reason": "Reading a registered buffer and processor save_pretrained file loss are unrelated." }, { "left": "issue:44514", "right": "issue:45103", "accept": false, - "reason": "Chat-template batching crash is unrelated to auto-docstring annotation parsing." + "reason": "Chat-template batching crash and auto_docstring annotation crash are different code paths." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "Layernorm epsilon precision issue is unrelated to remote-code/meta-init incompatibility." + "reason": "MLA layernorm epsilon precision mismatch and InternVL2 meta-init incompatibility are different bugs." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen/vision-related, but one is InternVL2 meta-init compatibility and the other is incorrect visual encoder keys on save_pretrained." - }, - { - "left": "issue:43723", - "right": "issue:44871", - "accept": false, - "reason": "Tokenizer loading failure and eos_token_id config mismatch are different bugs." - }, - { - "left": "issue:43723", - "right": "issue:44991", - "accept": false, - "reason": "Both are tokenizer loading issues, but they affect different models and likely different causes." + "reason": "Old InternVL2 meta-init issues and Qwen3.5 visual-key save_pretrained regressions are unrelated." } ] }, @@ -12058,8 +11952,6 @@ "issue:43066", "issue:43122", "issue:43232", - "issue:43408", - "issue:43475", "issue:43479", "issue:43576", "issue:43723", @@ -12082,8 +11974,10 @@ "issue:44568", "issue:44704", "issue:44743", + "issue:44871", "issue:44913", "issue:44945", + "issue:44991", "issue:45278", "issue:45290", "issue:45310", @@ -12092,7 +11986,9 @@ "issue:45753" ], "soft_pairs": [ + "issue:43723|issue:44871", "issue:30990|issue:38175", + "issue:43723|issue:44991", "issue:43723|issue:44442", "issue:41628|issue:43824", "issue:43479|issue:44913", @@ -12116,216 +12012,239 @@ "issue:44295|issue:44297", "issue:44704|issue:45290", "issue:42907|issue:44945", - "issue:43122|issue:44514", "issue:41628|issue:45278", + "issue:43122|issue:44514", "issue:43976|issue:44945", "issue:43784|issue:45600", "issue:39401|issue:44488", - "issue:43408|issue:43475", "issue:44261|issue:44373" ], - "prepared_review_unit_hash": "88a630e3b48200daffba5487d6b886504916e76f7a592a30189af05e795dde3f", + "prepared_review_unit_hash": "465037515db1d7074a19085e82715cb2cbb6e677b68bee01aac2d202febe3f65", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36536, - "estimated_input_tokens": 9134, - "estimated_eval_tokens": 18524 + "soft_pair_count": 32, + "serialized_chars": 36842, + "estimated_input_tokens": 9211, + "estimated_eval_tokens": 18678 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated tokenizer/import/config/model bugs. The soft pairs are mostly only broadly similar by subsystem, not the same concrete defect, so I would not merge any of them.", - "confidence": 0.86, - "canonical_issue_reason": "issue:43723 is the best umbrella issue here: it describes a broad tokenizer-loading regression in v5 and is the most representative of the repeated from_pretrained/tokenizer resolution failures.", + "summary": "This cluster is highly heterogeneous: most items are unrelated tokenizer/import/config/regression reports with only superficial similarity, so there is no strong duplicate core. The broadest representative is the v5 tokenizer-loading regression issue.", + "confidence": 0.41, + "canonical_issue_reason": "Issue 43723 is the broadest, most central tokenizer-loading regression in the set and overlaps the most with other tokenizer-related reports, though the cluster is still quite mixed.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43723 is the strongest single representative of the cluster because it is concrete, widely applicable, and closest to the recurring tokenizer import/load reports.", + "best_issue_reason": "Issue 43723 is the best single representative because it is broad, user-facing, and closest to the common tokenizer-loading theme seen in several nearby candidates.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Both are tokenizer-related, but one is a generic v5 loading failure and the other is a Gemma-3 EOS config mismatch; different bugs." + }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Different problems: Sentence Transformers loading hang vs. SigLIP2 zero probabilities." + "reason": "Completely different symptoms and code paths: Sentence Transformers loading hang vs zero probabilities in a SigLIP2 model." + }, + { + "left": "issue:43723", + "right": "issue:44991", + "accept": false, + "reason": "Both mention tokenizer loading in v5, but they are model-specific failures and do not look like the same underlying bug." }, { "left": "issue:43723", "right": "issue:44442", "accept": false, - "reason": "Both are tokenizer-loading failures, but the likely causes are too specific and not clearly the same bug without stronger evidence." + "reason": "Different tokenizer implementations and failure modes; no evidence they share the same concrete bug." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but for different symbols/classes and likely different export issues." + "reason": "Both are import errors, but for different missing symbols and different subsystems." }, { "left": "issue:43479", "right": "issue:44913", "accept": false, - "reason": "Both are config default/reload issues, but they affect different model configs and code paths." + "reason": "Unrelated config bugs: multimodal default config initialization vs GPTNeoX rotary_pct persistence." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "Different model-specific failures: FP8 accuracy vs. Qwen3.5-MoE from_pretrained error." + "reason": "Accuracy issue in FineGrainedFP8 vs a Qwen3.5 MoE from_pretrained error; no shared code-path." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "Unrelated: torch.split/indexer behavior vs. a docstring fix." + "reason": "One is a tensor-split indexing bug, the other is a docstring issue; not the same change or bug." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer behavior bugs, but on different models and different behaviors (offset mapping vs special-token insertion)." + "reason": "Both are tokenizer-related, but one is offset mapping in Qwen3 and the other is missing BOS/EOS insertion for mdeberta-v3." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "No common underlying defect; one is FP8 accuracy, the other is Qwen3-VL-MoE loading/compatibility." + "reason": "Different problems: numeric accuracy with FP8 vs Qwen3-VL-Moe loading/compatibility." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "Both mention import failures, but one is a specific huggingface_hub symbol and the other is a broad post-upgrade regression report." + "reason": "One specific missing import from huggingface_hub and one broad import-error report; too general to treat as the same bug." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Tokenizer offline-cache loading vs quantized model saving are unrelated." + "reason": "Offline tokenizer cache loading vs quantized model saving failure are unrelated." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation sync/cache bug vs D-FINE auxiliary-loss behavior are unrelated." + "reason": "Generation state update bug vs D-FINE auxiliary-loss behavior; different areas entirely." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Different model-family implementation issues; not the same code-path problem." + "reason": "Different model-specific architecture/config issues: RoPE layer handling vs tied-weights warning." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "Broken env CLI vs an import NameError in sentence-transformers integration are unrelated." + "reason": "Broken env CLI command vs import-time NameError in sentence-transformers; not the same bug." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "Different bugs in different model families; no shared fix path." + "reason": "FineGrainedFP8 accuracy issue and Qwen3_5 infinite loop are unrelated." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Offline tokenizer-cache loading and tokenizer decoder type regression are different defects." + "reason": "Both involve tokenizer/v5 behavior, but one is offline cache loading and the other is tokenizer decoder typing; different failures." }, { "left": "issue:43122", "right": "issue:44521", "accept": false, - "reason": "Different tokenizer/chat-template behaviors and different model surfaces." + "reason": "Both involve tokenization/chat templates, but one is version-dependent tokenization drift and the other is multimodal assistant mask generation." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 settings warning vs. torch.split/indexer behavior are unrelated." + "reason": "TF32 settings warning vs a torch.split indexing issue; unrelated." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "Import error for transformers.onnx vs tokenizer spacing after added token are unrelated." + "reason": "No module import error for transformers.onnx vs extra-space tokenizer behavior; unrelated." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "Both touch from_pretrained plumbing, but one is tokenizer mapping resolution and the other is processor kwargs forwarding." + "reason": "Both are from_pretrained-related, but one is TOKENIZER_MAPPING_NAMES handling and the other is AutoProcessor kwargs forwarding." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "Sentence Transformers loading hang vs. MacOS bus error with CLIP model are different failures." + "reason": "Loading hang in sentence-transformers vs MacOS bus error in a CLIP model; not the same bug." }, { "left": "issue:43232", "right": "issue:44743", "accept": false, - "reason": "Both involve generation/state handling, but they are model-specific and not the same concrete bug." + "reason": "Both touch generation/cache behavior, but the concrete failures differ: sync_gpus kwargs update vs recurrent-state reset in modular_qwen3_5." }, { "left": "issue:44295", "right": "issue:44297", "accept": false, - "reason": "Buffer access error vs tokenizer save metadata mismatch are unrelated." + "reason": "Reading a bufferized position_ids tensor vs tokenizer_class mismatch in saved config; unrelated." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Processor kwargs forwarding vs chat-template crash with tool calls are different code paths." + "reason": "AutoProcessor kwargs propagation vs chat_template crash with tool-call assistant messages; different code paths." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Quantized model save failure vs incorrect output under pipeline parallelism are unrelated." + "reason": "Dequantized model saving failure vs incorrect output under pipeline parallelism; unrelated." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Specific missing AutoImageProcessor import vs broad import-error report; not the same concrete issue." }, { "left": "issue:43122", "right": "issue:44514", "accept": false, - "reason": "General tokenization differences vs a multimodal apply_chat_template batching crash are not the same bug." + "reason": "Both are chat/tokenization issues, but one is version tokenization drift and the other is a batched apply_chat_template crash." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility problem vs pipeline-parallelism output bug; unrelated." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "Import NameError in downstream integration vs broken auto_mappings references are unrelated." + "reason": "Import NameError in sentence-transformers vs CI break from removed auto_mappings config references; different failures." }, { "left": "issue:39401", "right": "issue:44488", "accept": false, - "reason": "Tokenizer offset mapping bug vs model loading failure are only loosely related." + "reason": "Both are tokenizer loading-related, but the affected models and failure modes are different." }, { - "left": "issue:43408", - "right": "issue:43475", + "left": "issue:44261", + "right": "issue:44373", "accept": false, - "reason": "Same model family, but the warning about model type mismatch and the missing attribute error are different defects." + "reason": "Precision/config issue in MLA q_a_layernorm vs a docstring correction; not duplicates." }, { - "left": "issue:44261", - "right": "issue:44373", + "left": "issue:44514", + "right": "issue:44521", "accept": false, - "reason": "Precision/config problem vs documentation typo are unrelated." + "reason": "Both involve apply_chat_template, but one is a batched-padding crash and the other is zero assistant masks for multimodal inputs; distinct bugs." } ] }, @@ -12355,7 +12274,6 @@ "issue:36296", "issue:38617", "issue:39401", - "issue:42222", "issue:42371", "issue:42673", "issue:42757", @@ -12365,8 +12283,9 @@ "issue:43116", "issue:43122", "issue:43278", - "issue:43441", + "issue:43408", "issue:43452", + "issue:43475", "issue:43493", "issue:43673", "issue:43784", @@ -12399,6 +12318,7 @@ "issue:45753" ], "soft_pairs": [ + "issue:43408|issue:43475", "issue:42907|issue:43937", "issue:44360|issue:44485", "issue:44625|issue:44829", @@ -12429,229 +12349,228 @@ "issue:42371|issue:43874", "issue:43064|issue:43673", "issue:43976|issue:45446", - "issue:43116|issue:43493", - "issue:42222|issue:43441" + "issue:43116|issue:43493" ], - "prepared_review_unit_hash": "59529ab59610457af270b4719a2ec3be1bca8c78d189339256774c99f5c40224", + "prepared_review_unit_hash": "57afcd8047dc4ffa57986e6bc57645097e1888a68b534d60563f8a93fe19c1f8", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36896, - "estimated_input_tokens": 9224, - "estimated_eval_tokens": 18704 + "serialized_chars": 36958, + "estimated_input_tokens": 9240, + "estimated_eval_tokens": 18736 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is mostly a grab bag of distinct Transformers bugs/regressions, with only a few loose thematic similarities around v5 import/tokenizer issues. I would not merge any of the soft pairs as duplicates based on the available evidence.", - "confidence": 0.78, - "canonical_issue_reason": "issue:45278 is the broadest and most representative issue in the set: it captures the general post-upgrade import-regression theme that overlaps with several other reports, even though the other issues are more specific.", + "summary": "This cluster is mostly heterogeneous: the soft pairs share broad themes like imports, tokenization, or Qwen/SAM/GLM models, but they do not appear to be the same concrete bug or change. I would keep them separate; no PRs are present.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45278 is the broadest umbrella for the import-regression subset (many import errors after updating to 5.5.0), making it the best representative issue in this mixed cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best single issue for triage because it is an umbrella report, recent, and open; it can serve as the main pointer for related import-breakage reports.", + "best_issue_reason": "issue:45278 best captures the most central recurring theme here: upgrade-related import failures across multiple reports. It is still not a duplicate of the others, just the most useful anchor issue.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Same model family, but different bugs: a model-type warning vs a missing attribute on Sam3VisionEncoderOutput." + }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Different failures in different model paths: dequantized save/export for Ministral/Devstral vs invalid GenerationConfig for GLM-5." + "reason": "Different failures: dequantized save_pretrained vs GLM-5 GenerationConfig validation." }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "Unrelated problems: DSA indexer/ReLU discussion vs GLM-5 RoPE implementation." + "reason": "Unrelated: DSA indexer ReLU discussion vs GLM-5 RoPE implementation." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "Different bugs in Qwen3.5 config propagation vs flash-attention-3 training degeneration." + "reason": "Different code paths: Qwen3.5 num_labels propagation vs flash_attention_3 degenerate training." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Tokenizer offset_mapping bug for Qwen3 is not the same as multimodal assistant_masks being all zero in apply_chat_template." + "reason": "Tokenizer offset_mapping bug vs multimodal apply_chat_template assistant mask bug." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Separate issues: empty multi-label outputs in an example script vs Gemma 4 mm_token_type_ids defaults." + "reason": "Example-script multi-label output bug vs Gemma 4 mm_token_type_ids defaulting issue." }, { "left": "issue:29127", "right": "issue:30990", "accept": false, - "reason": "Different subsystems and symptoms: LayoutLMv3 error messaging vs Sentence Transformers loading stall." + "reason": "LayoutLMv3 error-message clarity vs Sentence Transformers loading hang; not the same defect." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe loading bug is not the same as tokenizer.save_pretrained writing a mismatched tokenizer_class." + "reason": "Qwen3-VL-Moe regression vs tokenizer save_pretrained metadata mismatch." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Different model-specific loading/saving failures; no shared code-path is evident." + "reason": "Different model-loading/saving issues: dequantized save failure vs Phi-4 multimodal load failure." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and likely different code paths; not enough to treat as the same bug." + "reason": "Both are tokenizer regressions, but they concern different models and likely different root causes." }, { "left": "issue:44038", "right": "issue:44295", "accept": false, - "reason": "One is a Qwen3-VL-Moe bug, the other is a position_ids buffer/read issue; unrelated symptoms." + "reason": "Different bugs: Qwen3-VL-Moe regression vs position_ids buffer access." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor-parallel training bug and example-script multi-label output bug are unrelated." + "reason": "Tensor-parallel training bug vs classification example returning empty results." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Saving dequantized Ministral/Devstral weights is unrelated to gguf_file breaking AutoTokenizer/AutoModelForCausalLM loading." + "reason": "Different loading/saving paths: dequantized save_pretrained vs gguf_file loading failure." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but for different missing symbols from different modules; not the same underlying bug." + "reason": "Both import errors, but for different symbols and causes; not the same underlying bug." }, { "left": "issue:44038", "right": "issue:44222", "accept": false, - "reason": "Both mention moe/models, but one is a load/import problem and the other is FP8 save_pretrained behavior; too different to merge." + "reason": "Same broad MoE area, but Qwen3-VL-Moe regression and FP8 save_pretrained are different issues." }, { "left": "issue:45216", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5 and version regressions, but one is a bad saved checkpoint and the other is a from_pretrained loading error; not clearly the same defect." + "reason": "Both involve Qwen3.5 regressions, but one is save_pretrained checkpoint corruption and the other is from_pretrained failure." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Distinct import failures for unrelated symbols: is_offline_mode vs HybridCache." + "reason": "Distinct import failures from different packages/symbols, not a shared root cause." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "Different models and behaviors: SigLIP2 implementation discrepancy vs Qwen3_5 image-specific infinite loop." + "reason": "Different model families and symptoms: SigLIP2 discrepancy vs Qwen3_5 image-specific infinite loop." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "Embedding dtype drift in train/eval is unrelated to incorrect outputs under pipeline parallelism." + "reason": "Dtype mismatch between train/eval vs incorrect output under pipeline parallelism." }, { "left": "issue:42907", "right": "issue:44222", "accept": false, - "reason": "Both concern model artifacts, but dequantized save failure and FP8 save_pretrained moe are different problems." + "reason": "Both mention save_pretrained, but dequantized save and FP8 MoE save are different bugs." }, { "left": "issue:43122", "right": "issue:44977", "accept": false, - "reason": "Tokenizer behavior change in v5 for generic same-tokenizer use is not clearly the same bug as Qwen3.5 flash-attention generation failure." + "reason": "Tokenizer behavior regression vs generation failure with flash-attention; different concrete problems." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "Missing embed_tokens.weight in UMT5Encoder loading is unrelated to AutoProcessor not forwarding kwargs to cached_file." + "reason": "Separate loading bugs: missing embed_tokens.weight vs cached_file kwargs not forwarded." }, { "left": "issue:43784", "right": "issue:45446", "accept": false, - "reason": "Sentence-transformers import NameError and flex_attention version-check bug are unrelated." + "reason": "Import-related, but one is a NameError in sentence-transformers and the other is a version check bug in flex_attention." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Wrong tokenizer decoder type and Deepseek tokenization regressions are both tokenizer-related, but not the same concrete bug." + "reason": "Different tokenizer regressions with different symptoms and likely different fixes." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "45278 is an umbrella for many import errors, but 38617 is a specific missing-symbol report; the evidence is too broad to call them duplicates." + "reason": "45278 is a broad import-regression report, but 38617 is a specific missing symbol; not the same concrete bug." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM multithreaded loading is unrelated to dtype mismatch in bfloat16 inference." + "reason": "VRAM leak in dataloader threads vs bfloat16 dtype mismatch in inference." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "GenerationConfig validation failure and GLM-5 RoPE implementation are separate issues." + "reason": "GenerationConfig validation issue vs RoPE implementation discussion." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "Different import/package problem vs LayoutLMv3 error-message clarity issue." + "reason": "No module named transformers.onnx vs LayoutLMv3 box-information error message; unrelated." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API warning/control issue is unrelated to missing get_number_of_image_patches in GLM46VImageProcessorFast." + "reason": "TF32 API/settings issue vs missing image-patch count method in GLM46VImageProcessorFast." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "FSDP2/PEFT optimizer-state bug and GenerationMixin cache missing during chunked_prefill affect different code paths." + "reason": "Different training/runtime failures: FSDP2 optimizer-state bug vs missing GenerationMixin cache in chunked prefill." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version compatibility break and flex_attention version-check bug are not the same underlying issue." + "reason": "Python-version compatibility issue vs incorrect PyTorch version check for AuxRequest import." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Example-script multi-label classification bug is unrelated to SigLIP2 implementation discrepancy." - }, - { - "left": "issue:42222", - "right": "issue:43441", - "accept": false, - "reason": "VitPose breakage and Ministral-3 FlashAttention failure are different model-specific regressions." + "reason": "Example-script classification bug vs SigLIP2 implementation discrepancy." } ] }, @@ -12670,8 +12589,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -12687,6 +12606,7 @@ "issue:43116", "issue:43408", "issue:43421", + "issue:43441", "issue:43450", "issue:43582", "issue:43742", @@ -12702,7 +12622,6 @@ "issue:44291", "issue:44410", "issue:44484", - "issue:44496", "issue:44779", "issue:44829", "issue:44933", @@ -12725,12 +12644,13 @@ "issue:45588" ], "soft_pairs": [ + "issue:42222|issue:43441", "issue:44075|issue:44936", "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:43827|issue:45446", @@ -12753,217 +12673,216 @@ "issue:42617|issue:45341", "issue:44933|issue:45278", "issue:42222|issue:45542", - "issue:42490|issue:42673", - "issue:43937|issue:44496" + "issue:42490|issue:42673" ], - "prepared_review_unit_hash": "1ebbbb66a9e6c9e0f8ab99bbeb68a27d2427c0b8f671f0eef040a5962069b739", + "prepared_review_unit_hash": "5d0bdbcdb34ece144caccd9921d5e6a848e590a8662f160eaa2d39da66a95cb7", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36519, - "estimated_input_tokens": 9130, - "estimated_eval_tokens": 18516 + "serialized_chars": 36465, + "estimated_input_tokens": 9117, + "estimated_eval_tokens": 18490 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are unrelated regressions across tokenizers, model loading, training, video processing, and import/runtime issues. None of the soft pairs look like the same concrete bug/change closely enough to merge as duplicates.", - "confidence": 0.82, - "canonical_issue_reason": null, + "summary": "These are mostly separate regression reports across different models, tokenizer paths, and runtime subsystems. I do not see any soft pair that clearly describes the same underlying bug, so I would not merge any of them as duplicates.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45278 is the broadest umbrella report in the set: a current open issue about widespread import failures after upgrading to v5.5.0, which best matches the cluster\u2019s import-regression theme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the broadest, most representative item in the set (import errors after upgrading), but it still only covers one subset of the cluster and not the full mix of unrelated reports.", + "best_issue_reason": "issue:45278 is the most suitable representative issue because it is open, broad, and could subsume related import-error regressions better than the highly specific model- or API-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "Different failures: VitPose model loading breakage vs Ministral-3 FlashAttention incompatibility." + }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "Different code paths and symptoms: optimizer argument handling vs trainer evaluate-after-train failure." + "reason": "Unrelated training bugs: SGD args ignored vs trainer.evaluate() failing after train()." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "One is multi-label classification output in an example script; the other is a v5 pipeline deprecation/error-message issue." + "reason": "Different concerns: multi-label example returning empty results vs pipeline() error text about unsupported translation tasks." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "Both are Qwen3.5-related, but one is save_pretrained checkpoint serialization and the other is from_pretrained loading; different bugs." + "reason": "Same model family, but different code paths: save_pretrained key layout vs from_pretrained loading error." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Same family name, but different model variants and different failure modes; not clearly the same underlying loader bug." + "reason": "Both mention Qwen/MoE, but the model/version and failure mode differ; not the same concrete bug." }, { - "left": "issue:43937", - "right": "issue:45563", + "left": "issue:45083", + "right": "issue:45446", "accept": false, - "reason": "GenerationConfig validation for GLM-5 is unrelated to a stale warning about num_return_sequences in paged generate()." + "reason": "Completely different subsystems: qwen3_omni_moe helper logic vs flex_attention import version check." }, { - "left": "issue:45083", - "right": "issue:45446", + "left": "issue:43937", + "right": "issue:45563", "accept": false, - "reason": "Completely different subsystems: feature-extractor length helper vs flex_attention PyTorch version check." + "reason": "GenerationConfig validation error is unrelated to the stale warning for num_return_sequences." }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping fallback bug and flash_attention sink-less-model crash are unrelated code paths." + "reason": "Tokenizer mapping None bug vs flash_attention crash on s_aux=None are unrelated." }, { "left": "issue:33453", "right": "issue:39401", "accept": false, - "reason": "Both are tokenizer-related, but one is loading regression and the other is incorrect offset_mapping for a specific tokenizer." + "reason": "Both tokenizer-related, but one is loading regression and the other is offset_mapping correctness for Qwen3." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Docs/pipeline removal issue is unrelated to a flex_attention import/version-check bug." + "reason": "Docs referencing pipeline() vs a PyTorch version check bug are unrelated." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Different tokenizer regressions on different models; too broad to treat as the same defect." + "reason": "Different tokenizer regressions affecting different models and symptoms." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "GLM generation-config validation and Qwen3.5 flash-attention generation failure are different bugs." + "reason": "Invalid GenerationConfig vs flash-attention generation problems are different bugs." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "One is a VRAM leak in multi-threaded dataloaders; the other is an isolated reproduction of a different earlier issue." + "reason": "VRAM leak in dataloader threads vs an isolated reproduction issue; no shared concrete defect." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "Unexpected init_empty_weights argument and PretrainedConfig type-checking breakage are separate API regressions." + "reason": "init_empty_weights argument handling vs PretrainedConfig type checking are separate code-path regressions." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Batched video processor shape bug and missing-torchvision AutoProcessor crash are related to video, but not the same concrete issue." + "reason": "Video processor batch-shape bug vs missing torchvision crash when using AutoProcessor are different failures." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "Dependency-version bump policy and a flex_attention import guard are unrelated." + "reason": "PEFT version bump issue and flex_attention import check are unrelated." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "A generic testing_utils bug and an xdist race in captured_info.txt are not the same underlying problem." + "reason": "A testing_utils bug and a pytest-xdist race in patched testing utils are not the same issue." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Tokenizer mapping None-handling and Mistral regex patch crash are distinct tokenizer bugs." + "reason": "Tokenizer mapping fallback bug vs Mistral regex patch crash are different tokenizer paths." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but for different missing symbols and different modules." + "reason": "Both are import errors, but they involve different symbols and different underlying regressions." }, { "left": "issue:39401", "right": "issue:43421", "accept": false, - "reason": "Offset mapping correctness and runtime special-token post-processor updates are separate tokenizer behaviors." + "reason": "Qwen3 offset_mapping bug vs runtime special-token/post-processor update issue are not the same defect." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen3, but one is missing projections in a layer and the other is a from_pretrained loading error." + "reason": "Missing projections in qwen3next is unrelated to Qwen3.5 from_pretrained loading failure." }, { "left": "issue:44484", "right": "issue:45310", "accept": false, - "reason": "Shard-size default discussion is unrelated to Qwen3.5 MoE loading failures." + "reason": "save_pretrained shard-size question vs from_pretrained error are not duplicates." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "Flash-attention training degeneration and incorrect save_pretrained checkpoint contents are different bugs." + "reason": "FlashAttention3 training degeneration is unrelated to incorrect checkpoint saving." }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 settings guidance and missing rms_norm_eps in MLA config are unrelated." + "reason": "TF32 API guidance issue vs MLA rms_norm_eps precision bug are different topics." }, { "left": "issue:42673", "right": "issue:43582", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM threads and Apple Silicon TypeError in caching_allocator_warmup are unrelated." + "reason": "VRAM leak in threads vs Apple Silicon TypeError in caching_allocator_warmup are unrelated." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "SAM3 model-type warning and MobileLLM key error are separate model-loading issues." + "reason": "Model-type warning for sam3 vs key error loading MobileLLM are unrelated." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "3d_parallel.py execution failure and a testing_utils bug are unrelated." + "reason": "3d_parallel.py execution failure and a testing_utils bug do not share the same code path." }, { "left": "issue:44933", "right": "issue:45278", "accept": false, - "reason": "A missing image_utils import is not the same as a broad collection of import errors after upgrade." + "reason": "Related only by broad import-error theme; 45278 is an umbrella report, not the same concrete bug as the image_utils import issue." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "VitPose breakage and missing-TensorFlow/tensorboard backend error are unrelated." + "reason": "VitPose breakage and a TensorFlow backend error from tensorboard-only installs are unrelated." }, { "left": "issue:42490", "right": "issue:42673", "accept": false, - "reason": "Local-model trailing-slash import warning and Qwen3ForCausalLM VRAM leak are different failures." - }, - { - "left": "issue:43937", - "right": "issue:44496", - "accept": false, - "reason": "Invalid GenerationConfig for GLM-5 and unrecognized OLMo-Hybrid model loading are unrelated." + "reason": "Trailing-slash local model path warning vs VRAM leak in dataloader threads are different bugs." } ] }, @@ -12982,8 +12901,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -12995,7 +12914,6 @@ "issue:42757", "issue:42831", "issue:43012", - "issue:43064", "issue:43066", "issue:43097", "issue:43122", @@ -13008,6 +12926,7 @@ "issue:43673", "issue:43874", "issue:43931", + "issue:43937", "issue:43976", "issue:43986", "issue:44038", @@ -13016,13 +12935,13 @@ "issue:44220", "issue:44261", "issue:44263", - "issue:44315", "issue:44368", "issue:44410", "issue:44483", "issue:44484", "issue:44485", "issue:44493", + "issue:44496", "issue:44829", "issue:44987", "issue:45003", @@ -13037,6 +12956,7 @@ "issue:45656" ], "soft_pairs": [ + "issue:43937|issue:44496", "issue:42175|issue:42222", "issue:44483|issue:45464", "issue:38617|issue:44220", @@ -13045,20 +12965,20 @@ "issue:44075|issue:45656", "issue:43673|issue:45341", "issue:43012|issue:43577", - "issue:44484|issue:45478", "issue:43874|issue:44261", + "issue:44484|issue:45478", "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", "issue:43931|issue:44410", "issue:43335|issue:45071", - "issue:45003|issue:45341", "issue:43931|issue:44038", + "issue:45003|issue:45341", "issue:43097|issue:43452", "issue:43986|issue:45341", "issue:44117|issue:44987", @@ -13067,250 +12987,249 @@ "issue:43874|issue:44263", "issue:43976|issue:45071", "issue:43986|issue:45372", - "issue:43531|issue:45341", "issue:44829|issue:45245", + "issue:43531|issue:45341", "issue:38617|issue:42757", - "issue:44261|issue:44485", - "issue:43064|issue:44315" + "issue:44261|issue:44485" ], - "prepared_review_unit_hash": "8993cfa9d2708962c8e65acc49e54c7e455ab0fce0d1e34644711acb23832c1f", + "prepared_review_unit_hash": "038bff0d50e6d97c538201a87f7830d2f1570620b2515b4ef0e337d42638bd91", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38034, - "estimated_input_tokens": 9509, - "estimated_eval_tokens": 19274 + "serialized_chars": 37997, + "estimated_input_tokens": 9500, + "estimated_eval_tokens": 19256 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of Transformers v5/v5.4 regressions and model-specific bugs that only look similar at the title level. I did not find any soft pair that is clearly the same underlying bug/change, so I rejected them all.", - "confidence": 0.9, - "canonical_issue_reason": "issue:45071 is the broadest, most central regression report here: a core v5.4.0 break in PretrainedConfig type checking that fits the cluster\u2019s general API/config regression theme.", + "summary": "This cluster is noisy and mostly contains unrelated issue reports that only look similar at the title level. I found no soft pair that clearly describes the same underlying bug/change, so I would not merge any of them.", + "confidence": 0.57, + "canonical_issue_reason": "Issue 45071 is the broadest, most central Transformers-v5 regression in the set and best represents the cluster\u2019s general compatibility-breakage theme.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45071 is the best representative issue because it is systemic, release-wide, and closer to the shared root theme than the more model-specific reports.", + "best_issue_reason": "Issue 45071 is the most reusable representative: it is a cross-cutting v5 regression, whereas most others are narrower model- or feature-specific failures.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GLM generation-config validation vs model-type/config loading failure; different bugs." + }, { "left": "issue:42175", "right": "issue:42222", "accept": false, - "reason": "Packaging/backend install issue vs. a VitPose model breakage; different symptoms and code paths." + "reason": "Backend installation packaging issue vs VitPose model breakage; unrelated." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Both involve chat/completions, but one is a general API request rejection and the other is a Qwen3.5-0.8B streaming failure; too different to merge." + "reason": "Both mention chat/completions, but one is a generic request rejection and the other is streaming inference on a specific model." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "Unrelated import error versus audio feature extraction bug." + "reason": "ImportError for a config symbol vs audio feature extraction issue; no concrete overlap." }, { "left": "issue:42371", "right": "issue:42831", "accept": false, - "reason": "TF32 settings guidance and FineGrainedFP8 accuracy are different issues." + "reason": "TF32 settings warning is unrelated to FineGrainedFP8 accuracy." }, { "left": "issue:43066", "right": "issue:43122", "accept": false, - "reason": "Both tokenizer-related, but one is decoder type metadata and the other is changed tokenization behavior; not the same bug." + "reason": "Tokenizer decoder-type regression vs tokenization output change; related area, not same bug." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "Different optimizer problems: unused SGD args vs. duplicate optimizer stepping under deepspeed." + "reason": "SGD argument handling vs duplicated optimizer step under deepspeed; different failure modes." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Generation cache regression and a testing_utils bug are unrelated." + "reason": "Generation cache regression vs a testing-utils bug; unrelated." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "One is a compile-time warning for bfloat16 models; the other is dtype not propagating into BLIP-2 submodules." + "reason": "Precision warning during compile vs dtype not propagating on load; same theme, different code paths." }, { - "left": "issue:44484", - "right": "issue:45478", + "left": "issue:43874", + "right": "issue:44261", "accept": false, - "reason": "save_pretrained shard-size question versus a Qwen3.5-MoE load error; no shared underlying defect." + "reason": "Missing image-patch helper vs MLA layernorm epsilon precision issue; unrelated." }, { - "left": "issue:43874", - "right": "issue:44261", + "left": "issue:44484", + "right": "issue:45478", "accept": false, - "reason": "Different multimodal/model-implementation bugs in unrelated model families." + "reason": "Save/sharding question vs Qwen3.5 MoE from_pretrained failure; not the same bug." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Qwen3Next missing projections and Qwen3.5-MoE from_pretrained failure are different model-specific failures." + "reason": "Missing projections in one model family vs a different Qwen3.5 MoE loading error." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "Removed tie_embeddings behavior and unexpected position-id keys are not the same bug." + "reason": "Removed tie-embeddings API vs unexpected position-id key warnings; different regressions." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "GLM tensor-split indexing and GLM-5 RoPE implementation are distinct defects." + "reason": "torch.split return behavior vs RoPE implementation; too different." }, { - "left": "issue:43012", - "right": "issue:43408", + "left": "issue:41628", + "right": "issue:45071", "accept": false, - "reason": "Completely different model families and failure modes." + "reason": "AutoImageProcessor import failure vs PretrainedConfig type-checking regression; different API breaks." }, { - "left": "issue:41628", - "right": "issue:45071", + "left": "issue:43012", + "right": "issue:43408", "accept": false, - "reason": "AutoImageProcessor import error and PretrainedConfig type checking breakage are unrelated." + "reason": "bfloat16 precision warning vs SAM3 model-type mismatch warning; unrelated." }, { "left": "issue:42222", "right": "issue:45459", "accept": false, - "reason": "VitPose model breakage versus tokenizer error masking from missing protobuf are different issues." + "reason": "VitPose model breakage vs tokenizer error masking without protobuf; no shared bug." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "FlashAttention regressions, but in different models and with different consequences; too broad to treat as one bug." + "reason": "Both involve flash attention, but different models and symptoms; not mergeable as one fix." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "Core config type checking breakage versus missing TensorFlow backend detection are unrelated." + "reason": "Config type checking regression vs TensorBoard-only backend issue; unrelated." }, { "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Different Qwen model-family loading problems with different symptoms." + "reason": "Qwen3-VL weight-shape mismatch vs qwen3next missing projections; different concrete problems." }, { "left": "issue:43335", "right": "issue:45071", "accept": false, - "reason": "SwitchTransformers sparse-layer construction bug and PretrainedConfig type checking are not the same defect." + "reason": "SwitchTransformers sparse-layer construction bug vs generic config type-checking regression." }, { - "left": "issue:45003", - "right": "issue:45341", + "left": "issue:43931", + "right": "issue:44038", "accept": false, - "reason": "Unsafe sys.modules access and a testing_utils bug are unrelated." + "reason": "Different Qwen3-VL model-loading failures; no evidence of the same root cause." }, { - "left": "issue:43931", - "right": "issue:44038", + "left": "issue:45003", + "right": "issue:45341", "accept": false, - "reason": "Both concern Qwen3 VL/MoE models, but the failures are different enough that they do not look like one underlying bug." + "reason": "Unsafe sys.modules access vs testing-utils bug; unrelated." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "Embedding tie behavior and gguf_file tokenizer/model loading failures are unrelated." + "reason": "Embedding-tie deprecation vs gguf_file tokenizer/model loading breakage; different issues." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "Video-model processor crash versus testing_utils bug; no shared root cause." + "reason": "AutoProcessor/torchvision crash vs testing-utils bug; unrelated." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Tokenizer mapping None-handling and a specific model load failure may be related at a high level, but not enough evidence for the same underlying bug." + "reason": "Tokenizer mapping None vs a specific model load failure; too little evidence of the same root cause." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "Completely different areas: processor loading versus save_pretrained sharding." + "reason": "Video AutoProcessor crash vs max_shard_size question; unrelated." }, { "left": "issue:43408", "right": "issue:44368", "accept": false, - "reason": "Model-type mismatch warning and LoRA tie_word_embeddings warning are separate warnings." + "reason": "Model-type warning for sam3_tracker vs embedding-tie warning in Qwen3.5 fine-tuning; different configs." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Different GLM-related bugs with no clear shared code-path." + "reason": "Image patch helper missing vs torch.split return values; unrelated." }, { "left": "issue:43976", "right": "issue:45071", "accept": false, - "reason": "Python-version compatibility issue versus config type checking regression." + "reason": "Python version support regression vs config type-checking regression; different breakages." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Video processor loading without torchvision and ReasoningEffort import failure are unrelated." + "reason": "Torchvision-dependent processor crash vs missing ReasoningEffort import; unrelated." }, { - "left": "issue:43531", - "right": "issue:45341", + "left": "issue:44829", + "right": "issue:45245", "accept": false, - "reason": "Qwen3-MoE sliding_window bug versus testing_utils bug." + "reason": "Flash-attention training degeneration vs category cardinality limit; unrelated." }, { - "left": "issue:44829", - "right": "issue:45245", + "left": "issue:43531", + "right": "issue:45341", "accept": false, - "reason": "FlashAttention_3 training degeneration and embedding category-limit error are unrelated." + "reason": "Qwen3-MoE sliding-window bug vs testing-utils bug; no overlap." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Different missing imports from different modules and dependency surfaces." + "reason": "Different import errors from different modules; not the same failure." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Different GLM bugs: missing rms_norm_eps handling versus RoPE implementation concerns." - }, - { - "left": "issue:43064", - "right": "issue:44315", - "accept": false, - "reason": "Distributed optimizer-state corruption and Liger Kernel application are separate training-path issues." + "reason": "LayerNorm epsilon precision issue vs RoPE implementation discussion; distinct bugs." } ] }, @@ -13365,6 +13284,7 @@ "issue:44062", "issue:44117", "issue:44222", + "issue:44315", "issue:44351", "issue:44368", "issue:44380", @@ -13376,7 +13296,6 @@ "issue:44683", "issue:44704", "issue:45103", - "issue:45290", "issue:45375", "issue:45478", "issue:45479", @@ -13384,12 +13303,13 @@ "issue:45753" ], "soft_pairs": [ + "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:44038|issue:44410", "issue:43502|issue:43992", "issue:43986|issue:44117", "issue:43097|issue:43873", @@ -13417,229 +13337,246 @@ "issue:43012|issue:44683", "issue:42757|issue:43824", "issue:43618|issue:44380", - "issue:38617|issue:43502", - "issue:44483|issue:45290" + "issue:38617|issue:43502" ], - "prepared_review_unit_hash": "de63ba080555bee251fe620993fcc6759d99f7294c56c6831c4239c834c770ae", + "prepared_review_unit_hash": "aa33f3466cdd8039ab38eb536d8dd3e1f429f0101b9aa9f0dad44c6c2ff10797", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38091, - "estimated_input_tokens": 9523, - "estimated_eval_tokens": 19302 + "serialized_chars": 38058, + "estimated_input_tokens": 9515, + "estimated_eval_tokens": 19286 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Loose v5-era Transformers regression cluster with mixed import errors, model-specific bugs, and API regressions; none of the soft pairs look like true duplicates.", - "confidence": 0.72, - "canonical_issue_reason": "issue:43097 is the broadest and most cluster-central v5 breaking-change report, with multiple soft-similarity links and some inbound attention, making it the closest hub even though the set is not a true duplicate cluster.", + "summary": "This is a loose similarity cluster of mostly unrelated Transformers issues: import errors, model-specific regressions, tokenizer/pipeline bugs, and training/runtime edge cases. I do not see any soft pair that is clearly the same concrete bug/change, so all soft edges should be rejected.", + "confidence": 0.91, + "canonical_issue_reason": "No single canonical issue fits the whole cluster; the items span unrelated bugs and regressions rather than one shared defect.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43097 is the best overall representative of the cluster's shared v5 regression theme, but it is still not a good duplicate anchor because nearby issues target different code paths.", + "best_issue_reason": "issue:43873 is the most active open item in the set and has broad user impact around quantized/offloaded loading, so it is the best representative anchor among these heterogeneous issues.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Different problems: distributed Trainer/FSDP2 optimizer-state corruption vs Liger Kernel application with model_init." + }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Different missing symbols and different import paths; both are import errors but not the same underlying bug." + "reason": "Both are import errors, but for different missing symbols and different code paths." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Both are model/config bugs, but they concern different Qwen variants and different missing fields/projections." + "reason": "Different model configs/fields: missing qwen3next projections vs missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "Tokenizer construction error vs mapping returning None; related area, but distinct failure modes and code paths." + "reason": "Tokenizer-related, but one is AddedToken argument duplication and the other is TOKENIZER_MAPPING_NAMES returning None." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "Config API removal vs flash-attention crash; unrelated bugs." + "reason": "Unrelated regressions: removed tie_embeddings_and_encoder_decoder vs flash_attention.py crashing on s_aux=None." }, { - "left": "issue:44038", - "right": "issue:44410", + "left": "issue:44704", + "right": "issue:45103", "accept": false, - "reason": "Both mention Qwen models, but one is a version/model support issue and the other is missing projections in a different architecture." + "reason": "Different utilities and failures: cached_file kwargs forwarding vs auto_docstring annotation handling." }, { - "left": "issue:44704", - "right": "issue:45103", + "left": "issue:44038", + "right": "issue:44410", "accept": false, - "reason": "AutoProcessor kwargs forwarding bug vs docstring annotation crash; different subsystems." + "reason": "Different Qwen model families and different failures; not the same underlying bug." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "Local-files-only network leak vs missing embed_tokens.weight on load; separate loading bugs." + "reason": "One is offline/local_files_only network leakage; the other is a missing embed_tokens.weight on load." }, { "left": "issue:43986", "right": "issue:44117", "accept": false, - "reason": "Missing torchvision path handling vs tokenizer mapping assumptions; different components." + "reason": "Different areas: AutoProcessor/video-model dependency handling vs tokenizer mapping None handling." }, { "left": "issue:43097", "right": "issue:43873", "accept": false, - "reason": "Config deprecation/removal issue vs quantization offloading behavior; not the same code-path problem." + "reason": "Different issues: API removal warning vs quantization/offloading behavior." }, { "left": "issue:43824", "right": "issue:43986", "accept": false, - "reason": "Different import/load failures involving different model families and dependencies." + "reason": "Both touch VL/processor loading, but one is an import failure and the other is a missing torchvision dependency crash." }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "No concrete evidence they are the same tokenizer bug; one is a decoder-type regression, the other is just an isolated reproduction of another issue." + "reason": "Wrong tokenizer decoder type in v5 is unrelated to the isolated reproduction issue." }, { "left": "issue:43097", "right": "issue:43901", "accept": false, - "reason": "API removal regression vs documentation mismatch for a pipeline argument; different issues." + "reason": "API removal vs documentation/behavior mismatch for return_all_scores; not the same defect." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "Chat/task support error message vs Pipeline save_pretrained failure; unrelated behaviors." + "reason": "Different pipeline failures: misleading task support message vs inability to save Pipeline objects." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing checkpoint weight vs shard-size question; not the same failure." + "reason": "Missing model weight loading vs a max_shard_size question; no shared underlying bug." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training blockers vs torch.compile SDPA failure in Bamba; different models and bugs." + "reason": "Different model/runtime bugs: xLSTM training blockers vs Bamba compile+SDPA failure." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "PyTorch warning about bfloat16 precision vs compile/runtime failure; not the same underlying problem." + "reason": "A compile warning on bfloat16 models is not the same as a compile failure on Bamba with SDPA." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "FlashAttention support gap for one model vs attention scaling ignored in backends for GPT2; separate issues." + "reason": "Different attention-related bugs in different models: unsupported Flash Attention 2.0 vs GPT2 attention scaling being ignored." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Compiler precision warning vs tie_word_embeddings warning; unrelated." + "reason": "Both mention warnings, but they concern unrelated behaviors and models." }, { "left": "issue:42886", "right": "issue:43381", "accept": false, - "reason": "Offline cache loading regression vs eval-mode gradient checkpointing restriction; different code paths." + "reason": "Offline cache loading failure vs gradient checkpointing in eval mode are unrelated." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Dequantized save_pretrained failure vs image-specific infinite loop; unrelated." + "reason": "Saving dequantized models and an infinite-loop inference bug are unrelated." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Saving dequantized weights vs saving Pipeline objects; both are save-related but not the same bug." + "reason": "Both mention saving, but one is dequantized model saving and the other is Pipeline.save_pretrained() support." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "Attention backend scaling bug vs sequence-classification zero-loss bug; different model behaviors." + "reason": "Attention-scaling backend bug vs degenerate zero-loss classification issue; no shared code-path." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Model-specific FlashAttention support vs Python version compatibility; unrelated." + "reason": "Different regressions: Flash Attention support for a model vs Python version compatibility." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Distributed optimizer-state corruption vs attention scaling; different subsystems." + "reason": "Distributed training optimizer-state corruption is unrelated to GPT2 attention scaling under SDPA/FlashAttention." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "xLSTM training issues vs Beit image processor label reduction bug; unrelated." + "reason": "Different model bugs: xLSTM training issues vs BeitImageProcessorFast label reduction behavior." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "OpenAPI/chat request rejection vs processor chat-template crash; similar domain, but different concrete failures." + "reason": "API request handling in chat/completions is unrelated to Qwen2_5_VLProcessor batching/padding crashes." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "Gemma3 config attribute error vs Apple Silicon allocator TypeError; unrelated." + "reason": "Gemma3Config missing vocab_size and an Apple Silicon allocator TypeError are unrelated." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "FP8 save_pretrained bug vs Qwen3.5 MoE from_pretrained error; both model-specific, but not the same code path." + "reason": "Same model family, but different directions and failures: FP8 save_pretrained vs from_pretrained error on a newer version." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Trailing-slash local path warning vs bfloat16 warning; not the same issue." + "reason": "Local path trailing-slash import warning is unrelated to the bfloat16 compile precision warning." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "Both are CLIP regressions, but one drops attentions and the other drops hidden_states; close family, different outputs and likely different fixes." + "reason": "Both are CLIP regressions, but one is missing attentions assignment and the other is hidden_states not returned; not the same concrete bug." }, { - "left": "issue:38617", - "right": "issue:43502", + "left": "issue:43012", + "right": "issue:44683", "accept": false, - "reason": "ImportError for a removed symbol vs unwanted network access under local_files_only; unrelated." + "reason": "Different compilation problems: a warning about precision vs a flex_attention failure on torch>=2.9." }, { - "left": "issue:44483", - "right": "issue:45290", + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Importing is_offline_mode from huggingface_hub is unrelated to a missing Qwen2_5_VLForConditionalGeneration import." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "Different model/code paths: CLIP output fields vs GPT2 attention scaling under backend attention implementations." + }, + { + "left": "issue:38617", + "right": "issue:43502", "accept": false, - "reason": "Chat/completions request handling vs assistant tool-call chat-template crash; different failure points." + "reason": "Distinct regressions: a missing configuration_utils symbol vs network calls despite local_files_only=True." } ] }, @@ -13703,8 +13640,8 @@ "issue:45071", "issue:45216", "issue:45278", + "issue:45290", "issue:45310", - "issue:45357", "issue:45375", "issue:45381", "issue:45440", @@ -13713,17 +13650,18 @@ "issue:45753" ], "soft_pairs": [ + "issue:44483|issue:45290", "issue:44038|issue:44230", + "issue:43756|issue:44485", "issue:44483|issue:45381", "issue:43299|issue:43867", - "issue:43756|issue:44485", "issue:43299|issue:45753", "issue:43901|issue:44655", "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:43066|issue:45381", "issue:36683|issue:44077", @@ -13744,235 +13682,222 @@ "issue:44355|issue:44855", "issue:36683|issue:43064", "issue:42890|issue:45561", - "issue:44938|issue:45071", - "issue:43867|issue:45357" + "issue:44938|issue:45071" ], - "prepared_review_unit_hash": "ac4630ae6b89a57168a21ce3a2420489bb5b734e0e5be67b3dc12eeec71af7ab", + "prepared_review_unit_hash": "890124d5fae0451df18f893dd9409c3eaf8da290a044502f5a168d2b43132670", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37162, - "estimated_input_tokens": 9291, - "estimated_eval_tokens": 18838 + "serialized_chars": 37171, + "estimated_input_tokens": 9293, + "estimated_eval_tokens": 18842 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly distinct regressions and feature bugs that only look similar at a high level (import/load/save/model version issues). I do not see any soft pair that is clearly the same underlying bug or change, so all soft edges should be rejected.", - "confidence": 0.88, - "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set: it covers widespread import breakage after upgrading from 4.57.0 to 5.5.0, which is the closest thing here to a cluster-level issue. It is still not a duplicate of the other items, just the best representative issue.", + "summary": "This cluster is a mixed bag of Transformers v5 regressions and model-specific bugs, not a single duplicate set. The shared wording mostly reflects broad themes like loading/import/config breakage, but the concrete failures differ too much for PR-style duplicate merging.", + "confidence": 0.96, + "canonical_issue_reason": "issue:45278 is the broadest umbrella here: it explicitly captures widespread import breakage after upgrading Transformers, which fits the cluster\u2019s overall regression pattern better than the more narrow model-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best global issue candidate because it is open, broad, and framed as a cross-cutting regression rather than a narrow model-specific bug. It can serve as the cluster anchor better than the many single-model or single-feature reports.", + "best_issue_reason": "issue:45278 is the best global representative because it is the most general and user-facing summary of the upgrade regression theme in this cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Different failures: chat/completions request acceptance vs apply_chat_template crashing on assistant tool-call messages." + }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Both mention Qwen3-VL-Moe, but one is a general loading bug and the other is fp8 inference support; different concrete problems." + "reason": "One is a Qwen3-VL-Moe loading bug; the other is fp8 inference support for Qwen3-VL/Qwen3.5 MoE, so the code paths and fixes differ." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Both mention RoPE, but one is a Smollm3 layer-selection bug and the other is GLM-5 RoPE implementation discussion; not the same defect." }, { "left": "issue:44483", "right": "issue:45381", "accept": false, - "reason": "API request handling for /v1/chat/completions is unrelated to Qwen2.5-VL video vision position IDs." + "reason": "Chat/completions request validation is unrelated to Qwen2.5-VL video vision_position_ids handling." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Different failure modes: Qwen3-VL-Moe loading breakage vs state_dict-sorted model load error." - }, - { - "left": "issue:43756", - "right": "issue:44485", - "accept": false, - "reason": "Different model families and bugs: Smollm3 RoPE layer dropping vs GLM-5 RoPE implementation discussion." + "reason": "Qwen3-VL-Moe loading regression and a generic sorted-state_dict loading error are different model-load problems." }, { "left": "issue:43299", "right": "issue:45753", "accept": false, - "reason": "Qwen3-VL-Moe loading failure and Qwen3.5 infinite-loop behavior on a specific image are not the same bug." + "reason": "Model loading breakage and an infinite-loop-on-specific-image report are unrelated symptoms and likely different fixes." }, { "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to pipeline save_pretrained serialization." + "reason": "Docs mismatch for return_all_scores is not the same as the pipeline save_pretrained serialization bug." }, { "left": "issue:36683", "right": "issue:43475", "accept": false, - "reason": "Gemma3Config missing vocab_size and Sam3VisionEncoderOutput missing fpn_position_embeddings are separate attribute regressions." + "reason": "Gemma3 missing vocab_size and Sam3VisionOutput missing fpn_position_embeddings are different model attribute regressions." }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Quantized model saving failure is not the same as inability to save Pipeline objects." + "reason": "Quantized model saving failure and pipeline object save_pretrained failure affect different save paths." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "FlashAttention 2 support gap for NemotronH is unrelated to GLM-5 GenerationConfig validation." + "reason": "Nemotron FlashAttention2 support gating and GLM-5 GenerationConfig validation are unrelated." }, { - "left": "issue:43867", - "right": "issue:45216", + "left": "issue:44077", + "right": "issue:44380", "accept": false, - "reason": "Both involve loading/saving, but one is a generic load error and the other is a Qwen3.5 save_pretrained checkpoint regression." + "reason": "One is a patchtsmixer post_init policy change; the other is GPT2 attention scaling being ignored under SDPA/FlashAttention." }, { - "left": "issue:44077", - "right": "issue:44380", + "left": "issue:43867", + "right": "issue:45216", "accept": false, - "reason": "PatchTSMixer post_init gating and GPT2 attention scaling under SDPA/FlashAttention are different code paths and bugs." + "reason": "Generic state_dict ordering load error and Qwen3.5 save_pretrained regression are different load/save bugs." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols and packages: transformers.generation vs huggingface_hub." + "reason": "Importing GenerationMixin from transformers.generation and importing is_offline_mode from huggingface_hub are distinct import regressions." }, { "left": "issue:43066", "right": "issue:45381", "accept": false, - "reason": "Tokenizer decoder type mismatch and Qwen2.5-VL video position IDs are unrelated regressions." + "reason": "Tokenizer decoder type in v5 and Qwen2.5-VL video position_ids are unrelated." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Gemma3Config attribute absence is unrelated to PatchTSMixer's post_init restriction." + "reason": "Gemma3 config attribute missing and patchtsmixer post_init restriction have no shared code path." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "base_model_prefix loading compatibility and pipeline save_pretrained are separate serialization problems." + "reason": "Base_model_prefix loading regression is not the same as pipeline serialization with save_pretrained." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both involve Python version compatibility, but they are different breakages in different versions and should not be merged." + "reason": "Python version compatibility regression and Python 3.14 load failure are too broad to treat as the same bug." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Gemma3Config missing vocab_size and Llama4Config missing pad_token_id are different missing-field bugs." + "reason": "Gemma3 missing vocab_size and Llama4 missing pad_token_id are separate config-attribute regressions." }, { "left": "issue:43352", "right": "issue:45593", "accept": false, - "reason": "FlashAttention 2 support for NemotronH is unrelated to D-FINE auxiliary loss behavior." + "reason": "FlashAttention2 support for Nemotron and D-FINE auxiliary-loss behavior are unrelated issues." }, { "left": "issue:44263", "right": "issue:45440", "accept": false, - "reason": "A torch.split return-value issue in GlmMoeDsaIndexer is not the same as DeepseekV3MoE divergence from remote implementation." + "reason": "torch.split return handling in GlmMoeDsaIndexer is not the same as DeepseekV3MoE diverging from remote implementation." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIPOutput attentions regression and Qwen3.5 save_pretrained checkpoint corruption are unrelated." + "reason": "Missing CLIPOutput attentions and Qwen3.5 save_pretrained corruption are different regressions." }, { "left": "issue:36683", "right": "issue:44242", "accept": false, - "reason": "Gemma3Config missing vocab_size is unrelated to MoE load-balancing loss omission when output_router_logits=False." + "reason": "Gemma3 vocab_size missing is unrelated to router-logit load-balancing loss behavior." }, { "left": "issue:44188", "right": "issue:44380", "accept": false, - "reason": "Both mention attention, but one is a torch.compile branch divergence and the other is backend-specific attention scaling being ignored." + "reason": "Torch.compile attention-kernel divergence and GPT2 attention scaling ignored in SDPA/FlashAttention are different backend bugs." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Distributed optimizer-state corruption and a typo in cache strategies are completely different." + "reason": "Distributed optimizer-state corruption and a cache-strategy typo are not the same problem." }, { "left": "issue:43352", "right": "issue:43931", "accept": false, - "reason": "NemotronH FlashAttention2 support and Qwen3-VL-30B-A3B-Instruct weight shape mismatch are unrelated." + "reason": "FlashAttention2 unsupported for Nemotron and Qwen3-VL weight-shape mismatch are unrelated model-loading/runtime failures." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "no_grad decoration for cache updates is unrelated to broad import failures after upgrade." + "reason": "no_grad decoration on cache updates is unrelated to broad import errors after upgrading to v5.5.0." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "UMT5Encoder missing embed_tokens.weight and Qwen3.5 MoE from_pretrained error are different model loading issues." + "reason": "UMT5 missing embed_tokens.weight and Qwen3.5 MoE from_pretrained errors are different model-load regressions." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "A sorted-state_dict load error is not the same as a missing Qwen3.5 vision config field being dropped by strict serialization." + "reason": "Generic sorted-state_dict loading error and missing deepstack_visual_indexes field are distinct issues." }, { "left": "issue:41669", "right": "issue:44246", "accept": false, - "reason": "Import * performance cleanup and intermittent import slowness are too broad and not clearly the same underlying bug." - }, - { - "left": "issue:44355", - "right": "issue:44855", - "accept": false, - "reason": "Running compiled Python files and a Python 3.13 IndentationError from torch.jit.script parsing are different failures." - }, - { - "left": "issue:36683", - "right": "issue:43064", - "accept": false, - "reason": "Gemma3Config attribute error and FSDP2/PEFT optimizer-state corruption are unrelated." + "reason": "44246 is a vague slow-import report, but 41669 is a specific import-* removal/performance fix; insufficient evidence they are the same concrete change." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Missing set_seed in tests and xdist races on captured_info.txt are separate test-infra issues." + "reason": "SAM HQ set_seed flakiness is unrelated to xdist races on captured_info.txt." }, { "left": "issue:44938", "right": "issue:45071", "accept": false, - "reason": "Python 3.14 load failure and PretrainedConfig type-checking regression are distinct version-specific problems." - }, - { - "left": "issue:43867", - "right": "issue:45357", - "accept": false, - "reason": "Generic load error when state_dict is sorted is not the same as incorrect visual encoder keys saved for Qwen3.5." + "reason": "Python 3.14 load failure and PretrainedConfig type-checking breakage are different release regressions, not one underlying bug." } ] }, @@ -13990,9 +13915,9 @@ "review_unit_index": 44, "review_unit_count": 47, "status": "reviewed", - "reason": null, - "source": "cache", - "cache_hit": true, + "reason": null, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -14023,7 +13948,6 @@ "issue:44077", "issue:44112", "issue:44222", - "issue:44315", "issue:44351", "issue:44355", "issue:44360", @@ -14038,6 +13962,7 @@ "issue:44977", "issue:45125", "issue:45341", + "issue:45357", "issue:45375", "issue:45381", "issue:45440", @@ -14046,6 +13971,7 @@ "issue:45850" ], "soft_pairs": [ + "issue:43867|issue:45357", "issue:43010|issue:45850", "issue:43992|issue:45478", "issue:43010|issue:43976", @@ -14066,216 +13992,239 @@ "issue:42890|issue:45341", "issue:44222|issue:45125", "issue:44387|issue:44556", - "issue:43010|issue:43065", "issue:42548|issue:44351", + "issue:43010|issue:43065", "issue:43502|issue:44485", "issue:44355|issue:44749", + "issue:45125|issue:45478", "issue:43010|issue:43232", "issue:30990|issue:31795", - "issue:45125|issue:45478", "issue:45125|issue:45375", "issue:30333|issue:33453", "issue:43299|issue:44556", - "issue:44315|issue:44360", - "issue:42831|issue:43874", - "issue:43502|issue:44222" + "issue:43502|issue:44222", + "issue:42831|issue:43874" ], - "prepared_review_unit_hash": "df923820a2b12abbb1e7acdd91f4f5f05e0871d3733040818934b3bb45c4b441", + "prepared_review_unit_hash": "53f14646eae41d8dcb897d8b34f53dca965f70842fb83ee84170c1736180031c", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 36988, - "estimated_input_tokens": 9247, - "estimated_eval_tokens": 18750 + "serialized_chars": 37013, + "estimated_input_tokens": 9254, + "estimated_eval_tokens": 18764 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly broad transformer-regression reports that only look superficially similar; the soft pairs are not strong duplicate matches, so I rejected them all. No PRs are present in the cluster.", - "confidence": 0.84, - "canonical_issue_reason": "Issue 44556 is the clearest canonical issue candidate: it describes a concrete, reproducible model-reload regression after upgrading from v4.57 to v5.2/v5.3, which is more actionable than the vaguer umbrella complaint about many models breaking.", + "summary": "These are mostly unrelated issue reports, with a few thematically similar loading/regression bugs but no true duplicate-level matches among the soft pairs.", + "confidence": 0.66, + "canonical_issue_reason": "issue:43502 is the clearest, most actionable bug report in the set: it describes a concrete, user-visible regression with an obvious repro and fix target.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44556 is the best issue for duplicate triage because it has a specific failure mode, version range, and user-visible symptom, making it easier to anchor related regression reports.", + "best_issue_reason": "issue:43502 is the best issue anchor because it is specific, reproducible, and directly actionable rather than broad or model-specific.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Different failures: state_dict ordering during model loading vs incorrect visual encoder keys on save_pretrained." + }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "Different bugs: one is about decorating cache/layer update methods with no_grad, the other is a vague broad complaint about many models breaking in v5." + "reason": "One is a specific no_grad decoration request; the other is a broad complaint that many models break." }, { "left": "issue:43992", "right": "issue:45478", "accept": false, - "reason": "Different model families and failure points: missing embed_tokens.weight in UMT5Encoder versus a Qwen3.5 MoE from_pretrained error." + "reason": "Both are loading problems, but they target different models and different root causes." }, { "left": "issue:43010", "right": "issue:43976", "accept": false, - "reason": "One is a generation/cache code-path issue; the other is a Python version compatibility problem. Not the same bug." + "reason": "Unrelated: cache update decoration vs a Python-version compatibility failure." }, { "left": "issue:44938", "right": "issue:45381", "accept": false, - "reason": "A Python 3.14 load failure is unrelated to the Qwen2.5-VL video vision_position_ids bug." + "reason": "Different areas: Python 3.14 import/load failure vs Qwen2.5-VL video position-id behavior." }, { "left": "issue:43618", "right": "issue:44077", "accept": false, - "reason": "CLIPOutput attentions regression and patchtsmixer post_init policy are separate code paths and symptoms." + "reason": "Different model/API issues: CLIPOutput attentions vs patchtsmixer post_init policy." }, { "left": "issue:42490", "right": "issue:44075", "accept": false, - "reason": "Local path import warnings and ignored SGD optimizer args are unrelated issues." + "reason": "Local model path warning and SGD optimizer arguments are unrelated bugs." }, { "left": "issue:36296", "right": "issue:43575", "accept": false, - "reason": "Both mention tensor parallelism, but one is a training bug and the other is an OOM when loading a specific model; not the same underlying defect." + "reason": "Both mention tensor parallelism, but one is a generic training bug and the other is a specific OOM on loading a Qwen model." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer missing ReLU and DeepseekV3 remote divergence are different implementations and failure modes." + "reason": "Different code paths: DSA indexer activation issue vs DeepseekV3MoE implementation divergence." }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "ImportError for AutoModel without PyTorch and tokenizer loading regression are separate problems." + "reason": "ImportError for missing PyTorch and tokenizer loading regression are not the same defect." }, { "left": "issue:43867", "right": "issue:44484", "accept": false, - "reason": "State-dict sorting load error versus max_shard_size default behavior are unrelated." + "reason": "Model load failure from sorted state_dict is unrelated to the save_pretrained shard-size question." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "An attentions-field regression in CLIPOutput is unrelated to an AMP/CUDA crash in LwDetrImageLoss." + "reason": "CLIPOutput attentions assignment and AMP/CUDA loss crash are different bugs." }, { "left": "issue:43618", "right": "issue:44977", "accept": false, - "reason": "These are distinct failures: missing attentions assignment versus Qwen3.5 flash-attention generation behavior." + "reason": "Different subsystems and symptoms: CLIPOutput field regression vs Qwen3.5 flash-attention generation bug." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "A flaky test needing set_seed is not the same as a Phi-4 multimodal model loading failure." + "reason": "A flaky test due to missing set_seed is unrelated to a model-loading failure." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/test-related, but the failure modes and affected components differ." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test seeding issue and auto_mappings stale config references are separate problems." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "One concerns unintended network requests under local_files_only, the other is an AddedToken keyword-argument TypeError." + "reason": "Local_files_only network leakage and AddedToken keyword collision are unrelated." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Both are loading-related, but one is tokenizer loading in v5 while the other is checkpoint reload failure after upgrading versions; the concrete code paths differ." + "reason": "Both involve version changes and loading, but one is tokenizer loading and the other is checkpoint reload after upgrade." }, { "left": "issue:42890", - "right": "issue:44112", + "right": "issue:45341", "accept": false, - "reason": "A seed-related test flake and a stale device override CI failure are unrelated." + "reason": "A flaky model integration test is not the same as a generic testing_utils bug." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "FP8 save_pretrained behavior and missing _tp_plan for tensor parallelism are different model/configuration issues." + "reason": "FP8 save_pretrained behavior and missing tensor-parallel plan are different feature gaps." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "CUDA reserved-memory OOM under int4 quantization is not the same as a checkpoint reload regression after version upgrade." + "reason": "Int4 memory regression and checkpoint reload incompatibility are different issues." }, { - "left": "issue:43010", - "right": "issue:43065", + "left": "issue:42548", + "right": "issue:44351", "accept": false, - "reason": "Cache update no_grad decoration and a dummy Conv2d in Sam3PixelDecoder are unrelated." + "reason": "Both are import errors, but for different symbols and likely different export fixes." }, { - "left": "issue:42548", - "right": "issue:44351", + "left": "issue:43010", + "right": "issue:43065", "accept": false, - "reason": "Both are import/export complaints, but they target different top-level symbols and likely different fixes." + "reason": "No_grad on update methods and a dummy Conv2d in Sam3PixelDecoder are unrelated." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "Local-files-only network leakage and GLM-5 RoPE implementation are unrelated." + "reason": "local_files_only network requests and GLM-5 RoPE implementation are unrelated defects." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled Python-file errors and a v5.3 slowdown in filtering are not the same bug." + "reason": "Compiled-Python-file errors and post-upgrade inference slowdown are different problems." }, { "left": "issue:45125", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5/MoE, but one is missing tensor-parallel metadata and the other is a from_pretrained error; not enough evidence for a duplicate." + "reason": "Both are Qwen3.5 MoE-related, but one is missing tensor-parallel metadata and the other is a generic from_pretrained failure." }, { - "left": "issue:30333", - "right": "issue:33453", + "left": "issue:43010", + "right": "issue:43232", "accept": false, - "reason": "MLflow reporting on failed training jobs is unrelated to tokenizer loading regression." + "reason": "Generation state update timing and no_grad decoration are different code paths." }, { - "left": "issue:43299", - "right": "issue:44556", + "left": "issue:30990", + "right": "issue:31795", "accept": false, - "reason": "These are both version-related loading regressions, but they affect different models and symptoms; not the same underlying bug." + "reason": "Stuck model loading and documentation confusion are not the same bug." }, { - "left": "issue:44315", - "right": "issue:44360", + "left": "issue:45125", + "right": "issue:45375", "accept": false, - "reason": "Liger Kernel application with model_init and a DSA indexer activation issue are unrelated." + "reason": "Qwen3.5 tensor-parallel metadata and missing vision config field are separate configuration bugs." }, { - "left": "issue:42831", - "right": "issue:43874", + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLFlow reporting on failed jobs and tokenizer loading regression are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:44556", "accept": false, - "reason": "An FP8 accuracy issue and a missing image-patch method are different bugs." + "reason": "Both mention upgrade-related loading regressions, but they affect different models and failure modes." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "Local_files_only network requests and FP8 save_pretrained moe are unrelated." + "reason": "Network access despite local_files_only and FP8 save_pretrained behavior are not duplicates." } ] }, @@ -14310,7 +14259,6 @@ "issue:42371", "issue:42548", "issue:42757", - "issue:42890", "issue:43502", "issue:43525", "issue:43526", @@ -14325,6 +14273,7 @@ "issue:44230", "issue:44261", "issue:44273", + "issue:44315", "issue:44355", "issue:44360", "issue:44393", @@ -14349,13 +14298,14 @@ "issue:45850" ], "soft_pairs": [ + "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:33453|issue:34689", "issue:44222|issue:44230", + "issue:33453|issue:34689", "issue:42757|issue:43576", "issue:33453|issue:36296", "issue:43576|issue:43824", @@ -14378,261 +14328,254 @@ "issue:44273|issue:45003", "issue:44393|issue:45753", "issue:44393|issue:45071", - "issue:43576|issue:44556", "issue:44038|issue:45125", + "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:44704|issue:44908" ], - "prepared_review_unit_hash": "fd84916c165334b50e05d8f014f91b08a8202235ab14206372fbc14f1e199895", + "prepared_review_unit_hash": "f45d1b0f4a8716cd6529002bf46596050cd19ee182ccedd680a8836fc92f822d", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 39, - "serialized_chars": 38530, - "estimated_input_tokens": 9633, - "estimated_eval_tokens": 19522 + "serialized_chars": 38478, + "estimated_input_tokens": 9620, + "estimated_eval_tokens": 19496 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose mix of unrelated bug reports, regressions, import errors, docs issues, and model-specific failures. The soft links are mostly surface-level similarity around loading/importing/Transformers v5 behavior, but they do not appear to describe the same underlying defect.", - "confidence": 0.82, - "canonical_issue_reason": "Issue 43502 is the clearest, most actionable bug report in the set: a specific loading behavior regression with a concrete symptom and narrow scope.", + "summary": "The cluster is mostly a collection of unrelated Transformers v5-era regressions and import/loading complaints. Issue 45850 is the broadest umbrella, but most pairs are not true duplicates; they only share a generic theme like 'v5 broke something.'", + "confidence": 0.81, + "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set and best matches the overall 'v5 breaks many models' regression theme.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43502 is the best representative issue because it is precise, reproducible, and describes a concrete code-path bug rather than a broad regression claim or documentation complaint.", + "best_issue_reason": "issue:45850 is the best single anchor for the cluster, though it is too vague to serve as a precise duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Different subsystems and symptoms: model_init/Liger Kernel vs DSA indexer/ReLU." + }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "'API requests are made despite setting local_files_only=True' is a concrete loader bug; 'racoon' is not a comparable bug report and does not match the same defect." + "reason": "Unrelated issues: offline/local_files_only behavior vs an empty 'racoon' report." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Both mention failures after changes, but one is about compiled Python file loading and the other about PretrainedConfig type checking; these are different code paths." + "reason": "One is about compiled Python file loading; the other is PretrainedConfig type checking." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "Training job/MLflow reporting and tensor-parallel training are unrelated problems." + "reason": "MLFlow reporting on failed training jobs is unrelated to tensor-parallel training behavior." }, { "left": "issue:44355", "right": "issue:44655", "accept": false, - "reason": "Compiled-file errors and save_pretrained pipeline serialization failures are different bugs with no shared failing behavior." + "reason": "Compiled Python file errors and pipeline save_pretrained failures are different bugs." }, { "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "Docs still referencing pipeline() is a documentation issue; 'racoon' is not a matching technical defect." + "reason": "Docs mentioning pipeline() after v5 removals is unrelated to the 'racoon' issue." }, { - "left": "issue:33453", - "right": "issue:34689", + "left": "issue:44222", + "right": "issue:44230", "accept": false, - "reason": "Both are loading regressions, but one is tokenizer loading and the other is Llama 3.2 vision model loading; too broad to be the same bug." + "reason": "Both mention FP8/MoE, but one is save_pretrained behavior and the other is inference support." }, { - "left": "issue:44222", - "right": "issue:44230", + "left": "issue:33453", + "right": "issue:34689", "accept": false, - "reason": "Both are fp8/Qwen3-related, but one is save_pretrained for moe models and the other is inference support; different change requests." + "reason": "Both are loading regressions, but they point to different code paths and a specific Llama Vision case." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "ImportError from huggingface_hub and broken transformers env command are unrelated failures." + "reason": "ImportError from huggingface_hub is unrelated to the v5 env command breakage." }, { "left": "issue:33453", "right": "issue:36296", "accept": false, - "reason": "Tokenizer loading regression and tensor parallel training bug are separate problems." + "reason": "Tokenizer loading regression is unrelated to tensor-parallel training." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "Broken env command and missing Qwen2_5_VL symbol are different import/CLI issues." + "reason": "Broken env command and missing Qwen2.5-VL export are separate failures." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "CLI breakage in v5 and PretrainedConfig type-checking regression do not target the same code path." + "reason": "Env command breakage is unrelated to PretrainedConfig type-checking regressions." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Documentation confusion about forward args is unrelated to an ImportError for layer_type_validation." + "reason": "Documentation confusion in forward args is unrelated to an ImportError for layer_type_validation." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "'racoon' has no discernible technical content to match the Gemma audio positional encoding bug." + "reason": "A placeholder issue and Gemma audio positional encoding bug are not duplicates." }, { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "Cannot import AutoImageProcessor and cannot import PreTrainedModel are different symbol-export regressions." + "reason": "Both are import errors, but for different symbols and likely different export gaps." }, { "left": "issue:38617", "right": "issue:43576", "accept": false, - "reason": "Missing configuration_utils symbol and broken env CLI are unrelated." + "reason": "ImportError for layer_type_validation and env command breakage are different problems." }, { "left": "issue:43576", "right": "issue:44938", "accept": false, - "reason": "The env command issue and Python 3.14 load failure are distinct environment/runtime problems." + "reason": "Broken env command is not the same as Python 3.14 load failures." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "Generation cache missing during chunked prefill and a cache-strategy typo are not the same defect." + "reason": "Chunked_prefill cache issue and a cache-strategy typo are not the same bug." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs handling and 'racoon' are unrelated." + "reason": "Scheduler kwargs handling is unrelated to the placeholder 'racoon' issue." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to the non-technical 'racoon' issue." + "reason": "RoPE implementation discussion is unrelated to the placeholder issue." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor parallel training bug and TextClassificationPipeline docs mismatch are different areas and failure modes." + "reason": "Tensor-parallel training bug is unrelated to documentation about return_all_scores." }, { "left": "issue:41628", "right": "issue:44355", "accept": false, - "reason": "AutoImageProcessor import failure and compiled Python file loading errors are different bugs." + "reason": "AutoImageProcessor import error and compiled-file loading errors are different regressions." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor parallel training and BeitImageProcessorFast reduce_labels behavior are unrelated." + "reason": "Tensor parallel training and BeitImageProcessorFast label reduction are unrelated." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "Qwen3.5 chat crash and generic 'Since 5.0 version it breaks too many models' are too vague to establish the same underlying bug." + "reason": "Qwen3.5 chat crash is a specific model/runtime bug; 45850 is only a broad umbrella." }, { "left": "issue:44261", "right": "issue:44360", "accept": false, - "reason": "Missing rms_norm_eps in MLA q_a_layernorm and a DSA indexer ReLU complaint are not the same code-path problem." + "reason": "MLA rms_norm_eps precision issue is unrelated to the DSA/ReLU report." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "Missing processor.save_pretrained files and a generic bug report are not clearly the same issue." + "reason": "Missing processor.save_pretrained files is unrelated to the placeholder bug report." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Lazy loading failure and undefined tf backend error are unrelated initialization issues." + "reason": "Lazy loading problems and tensorboard/TensorFlow backend errors are different failures." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Lazy loading and unsafe sys.modules access are not the same defect." + "reason": "Lazy loading behavior and unsafe sys.modules access are related only broadly, not as the same concrete bug." }, { "left": "issue:44393", "right": "issue:45753", "accept": false, - "reason": "2D bounding-box hallucination/error and infinite loop on a specific image are both Qwen3-VL, but they are different symptoms and likely different bugs." + "reason": "Different Qwen model/version issues: 2D box output vs infinite loop on one image." }, { "left": "issue:44393", "right": "issue:45071", "accept": false, - "reason": "Model-specific Qwen3-VL output bug and PretrainedConfig type checking regression are unrelated." - }, - { - "left": "issue:43576", - "right": "issue:44556", - "accept": false, - "reason": "Broken env command and checkpoint reload incompatibility after upgrade are different problems." + "reason": "Qwen3-VL output bugs are unrelated to PretrainedConfig type-checking." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "Qwen3-VL-Moe bug and missing _tp_plan for tensor parallelism are separate model/infrastructure issues." - }, - { - "left": "issue:43525", - "right": "issue:45850", - "accept": false, - "reason": "Llama4Config pad_token_id attribute error and a generic 'many models break' complaint are not the same concrete bug." + "reason": "Generic Qwen3-VL-MoE bug report and missing _tp_plan for tensor parallelism are distinct." }, { - "left": "issue:41669", - "right": "issue:43673", + "left": "issue:43576", + "right": "issue:44556", "accept": false, - "reason": "Removing import * from models and GenerationMixin cache behavior are unrelated changes." + "reason": "Env command breakage is unrelated to checkpoint reload incompatibility after upgrade." }, { - "left": "issue:44556", - "right": "issue:45092", + "left": "issue:43525", + "right": "issue:45850", "accept": false, - "reason": "Checkpoint reload incompatibility after v5 upgrade and InternVL2 remote-code meta initialization incompatibility are both version-regression reports but different concrete failures." + "reason": "Missing pad_token_id in Llama4Config is too specific to be the same as the broad v5 breakage umbrella." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "Cache strategy typo and generic v5 breakage are not the same bug." + "reason": "Cache strategy typo is not the same as the broad 'v5 breaks too many models' complaint." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "StableLmConfig pad_token_idx regression and generic v5 breakage are too broad to merge." + "reason": "StableLmConfig pad_token_idx regression is a separate model-specific issue." }, { "left": "issue:42371", @@ -14644,13 +14587,7 @@ "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "AutoProcessor kwargs forwarding and inverse_sqrt scheduler kwargs handling affect different APIs and code paths." - }, - { - "left": "issue:42890", - "right": "issue:44908", - "accept": false, - "reason": "SamHQ test seeding flakiness and scheduler kwarg handling are unrelated." + "reason": "Missing kwargs in AutoProcessor.from_pretrained is a different code path from scheduler kwargs handling." } ] }, @@ -14669,8 +14606,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -14680,6 +14617,7 @@ "issue:37428", "issue:41669", "issue:42371", + "issue:42890", "issue:42913", "issue:43066", "issue:43352", @@ -14716,7 +14654,6 @@ "issue:45092", "issue:45230", "issue:45278", - "issue:45290", "issue:45310", "issue:45341", "issue:45468", @@ -14724,6 +14661,7 @@ "issue:45850" ], "soft_pairs": [ + "issue:42890|issue:44908", "issue:45278|issue:45850", "issue:44246|issue:44273", "issue:43408|issue:44315", @@ -14731,8 +14669,8 @@ "issue:43352|issue:44393", "issue:43576|issue:44908", "issue:44297|issue:45230", - "issue:44908|issue:45071", "issue:44295|issue:45230", + "issue:44908|issue:45071", "issue:41669|issue:44351", "issue:41669|issue:45341", "issue:44492|issue:44945", @@ -14750,8 +14688,8 @@ "issue:44273|issue:44908", "issue:43519|issue:44485", "issue:43519|issue:44908", - "issue:31795|issue:34689", "issue:41669|issue:43704", + "issue:31795|issue:34689", "issue:31795|issue:37428", "issue:44315|issue:45092", "issue:44485|issue:45468", @@ -14763,317 +14701,298 @@ "issue:44263|issue:45468", "issue:43606|issue:45468", "issue:44261|issue:45468", - "issue:43704|issue:45341", "issue:42913|issue:43066", - "issue:44908|issue:44998", - "issue:44861|issue:45290", - "issue:44492|issue:44998", - "issue:44998|issue:45850" + "issue:44908|issue:44998" ], - "prepared_review_unit_hash": "5c74ca2c4863133a75a48fbe90f6da2a1abea31c1235540da7af637a460c1b9d", + "prepared_review_unit_hash": "d2b11dbba0b97c79375e58757a758ab2c19c1709d569f65a3a1961b5a71158dd", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 45, - "serialized_chars": 40511, - "estimated_input_tokens": 10128, - "estimated_eval_tokens": 20512 + "soft_pair_count": 42, + "serialized_chars": 39635, + "estimated_input_tokens": 9909, + "estimated_eval_tokens": 20074 }, "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 a few nearby themes around v5 regressions, tokenizer behavior, import/loading problems, and model-specific bugs. I found no soft-edge pair that clearly describes the same underlying defect or change.", - "confidence": 0.74, - "canonical_issue_reason": "issue:45278 is the most concrete and broadly representative regression report in the set; it has a real user-facing failure mode and a clear version jump, unlike the many very specific or generic outliers.", + "summary": "This cluster is mostly a mix of Transformers v5 regression reports across tokenizers, imports, loading, and model-specific failures. issue:45278 remains the best umbrella issue, but the soft edges should stay conservative: none of these pairs are clearly the same concrete bug from the provided titles alone.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45278 is still the strongest canonical issue: it is open and explicitly captures the broad multi-model breakage after the 4.57.0 -> 5.5.0 upgrade, making it a practical umbrella for this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best single issue to anchor the cluster because it is specific enough to be actionable while still capturing the broad upgrade-regression theme seen in nearby reports.", + "best_issue_reason": "issue:45278 is the best global issue target because it is open, broad enough to subsume related v5 breakage reports, and specific enough to remain useful as a tracker.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "SamHQ test flakiness from missing set_seed is unrelated to inverse_sqrt scheduler kwargs handling." + }, { "left": "issue:45278", "right": "issue:45850", "accept": false, - "reason": "Both are broad upgrade-regression complaints, but they are too vague and not the same concrete bug or code path." + "reason": "Both are broad v5 breakage complaints, but they do not describe the same concrete failure mode." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Both mention import/loading, but one is about slow imports and the other about lazy loading failing; not clearly the same defect." + "reason": "Import slowness and lazy-loading failure are related themes, but not clearly the same bug." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "Different problems in different areas: model type mismatch versus Liger Kernel application in model_init." + "reason": "Model-type warning for sam3_tracker is unrelated to Liger Kernel application in model_init." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python version compatibility and Qwen3-VL bounding-box output are unrelated." + "reason": "Python version compatibility and Qwen3-VL bbox output errors are unrelated." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash-Attention support gating for Nemotron is unrelated to Qwen3-VL output errors." + "reason": "Flash Attention 2 support error is unrelated to Qwen3-VL bounding-box hallucination/output bugs." }, { "left": "issue:43576", "right": "issue:44908", "accept": false, - "reason": "Broken env command and scheduler kwargs handling are unrelated subsystems and failure modes." + "reason": "Broken transformers env command is unrelated to scheduler kwargs being ignored." }, { "left": "issue:44297", "right": "issue:45230", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch is too specific to match a generic bug report." + "reason": "Specific tokenizer_class mismatch is too different from a generic bug report to treat as the same issue." }, { - "left": "issue:44908", - "right": "issue:45071", + "left": "issue:44295", + "right": "issue:45230", "accept": false, - "reason": "Scheduler kwargs handling and PretrainedConfig type checking are different code paths." + "reason": "Position_ids buffer access error is not the same as an unspecified generic bug report." }, { - "left": "issue:44295", - "right": "issue:45230", + "left": "issue:44908", + "right": "issue:45071", "accept": false, - "reason": "Buffer handling for position_ids is unrelated to the generic bug report." + "reason": "Scheduler kwargs bug and PretrainedConfig type-checking regression are different code paths." }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-star cleanup and HybridCache import failure are not the same bug." + "reason": "Import-* performance regression is unrelated to the HybridCache import error." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Model import cleanup is unrelated to the testing_utils bug." + "reason": "Import-* performance regression is unrelated to a testing_utils bug." }, { "left": "issue:44492", "right": "issue:44945", "accept": false, - "reason": "A typo in cache strategy text is unrelated to incorrect LLM output under pipeline parallelism." + "reason": "Cache-strategy typo and pipeline-parallelism output corruption are unrelated." }, { "left": "issue:43874", "right": "issue:45850", "accept": false, - "reason": "A missing image-patch method in one processor is not the same as a broad complaint that many models break." + "reason": "Missing get_number_of_image_patches is a concrete GLM46V bug, not the same as generic model breakage." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Qwen3-VL bbox errors and Qwen3.5 MoE from_pretrained errors are different models and different failure modes." + "reason": "Qwen3-VL bbox issues and Qwen3.5 MoE from_pretrained errors are different failures." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Optional post_init policy and tied-weight key crashes are unrelated." + "reason": "patchtsmixer post_init policy and tied-weight-key AttributeError are unrelated." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Docs still referencing pipeline() and inverse_sqrt kwargs handling are unrelated." + "reason": "Docs still referencing pipeline() is unrelated to inverse_sqrt scheduler kwargs." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion for model.forward is unrelated to tokenizer loading regression." + "reason": "Confusing forward docs and tokenizer loading regression are different issues." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Different training/setup issues: model_init not applying Liger Kernel versus flash_attention_3 causing degenerate training." + "reason": "Liger Kernel not applied with model_init is unrelated to flash_attention_3 degenerate training." }, { "left": "issue:41669", "right": "issue:44908", "accept": false, - "reason": "Import-star cleanup does not match scheduler kwargs behavior." + "reason": "Import-* performance regression is unrelated to a scheduler keyword bug." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken env command and lazy loading failures are different issues." + "reason": "Broken env CLI and lazy-loading failure are only loosely related." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Saving pipelines and tied-weight key crashes are different bugs." + "reason": "Pipeline save_pretrained failure is unrelated to _tied_weight_keys list handling." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "DSA indexer ReLU discussion is unrelated to Gemma-4 audio positional encoding." + "reason": "DSA indexer ReLU issue and Gemma-4 audio positional encoding bug are different model-specific problems." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "A generic bug report does not make this the same underlying Qwen3.5 MoE from_pretrained issue." + "reason": "A generic bug report is too vague to confirm it is the same as the Qwen3.5 MoE from_pretrained error." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading problems and scheduler kwargs are unrelated." + "reason": "Lazy loading and inverse_sqrt scheduler kwargs are unrelated." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different model-specific math bugs." + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different model bugs." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Timestamp calculation in a processor is unrelated to scheduler kwargs." + "reason": "Timestamp calculation in Qwen3VL is unrelated to scheduler kwargs handling." }, { - "left": "issue:31795", - "right": "issue:34689", + "left": "issue:41669", + "right": "issue:43704", "accept": false, - "reason": "A forward-docs clarification issue is not the same as a concrete model loading regression." + "reason": "Import-* performance regression is unrelated to a VRAM leak in dataloader threads." }, { - "left": "issue:41669", - "right": "issue:43704", + "left": "issue:31795", + "right": "issue:34689", "accept": false, - "reason": "Import-star cleanup and VRAM leakage in dataloader threads are different concerns." + "reason": "Model.forward documentation confusion is unrelated to a Llama 3.2 model-loading regression." }, { "left": "issue:31795", "right": "issue:37428", "accept": false, - "reason": "Documentation confusion and flash-attention import failure are unrelated." + "reason": "Forward-doc confusion is unrelated to the flash-attention import error." }, { "left": "issue:44315", "right": "issue:45092", "accept": false, - "reason": "Both touch model creation/loading, but they are about different checkpoints and different failure causes." + "reason": "Both involve model initialization, but they describe different concrete incompatibilities." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Different models and different implementation areas; not the same bug." + "reason": "Different models and different bugs: GLM-5 RoPE versus Gemma-4 audio positional encoding." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak in dataloader threads is unrelated to RoPE implementation." + "reason": "VRAM leak in multi-threaded dataloading is unrelated to GLM-5 RoPE issues." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API guidance and VRAM leakage are unrelated." + "reason": "TF32 configuration guidance is unrelated to a VRAM leak." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "Unexpected API requests with local_files_only is unrelated to timestamp calculation." + "reason": "local_files_only network requests and Qwen3VL timestamp calculation are unrelated." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp math and save_pretrained missing files are different bugs." + "reason": "Timestamp calculation bug and processor.save_pretrained missing files are unrelated." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Timestamp calculation and VRAM leak are unrelated." + "reason": "Timestamp calculation bug and VRAM leak are unrelated." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "GlmMoeDsaIndexer split semantics and Gemma-4 audio positional encoding are unrelated." + "reason": "torch.split return-values issue and Gemma-4 audio positional encoding bug are unrelated." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch and audio positional encoding are unrelated." + "reason": "CPU offload device mismatch and Gemma-4 audio encoding issue are unrelated." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "Precision differences in MLA q_a_layernorm are unrelated to Gemma-4 audio positional encoding." - }, - { - "left": "issue:43704", - "right": "issue:45341", - "accept": false, - "reason": "VRAM leakage and a testing utility bug are unrelated." + "reason": "LayerNorm epsilon precision issue and Gemma-4 audio encoding issue are unrelated." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both are tokenizer/v5 related, but one is a broad behavioral change and the other is a specific decoder-type bug; not clearly the same defect." + "reason": "The titles point to two different tokenizer-v5 regressions: one is a broad behavior difference, the other is a specific wrong decoder-type bug. That is not enough to confirm the same underlying issue." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs and an unrelated generic issue title have no shared bug." - }, - { - "left": "issue:44861", - "right": "issue:45290", - "accept": false, - "reason": "Tied-weight key handling and apply_chat_template tool-call crashes are different code paths." - }, - { - "left": "issue:44492", - "right": "issue:44998", - "accept": false, - "reason": "Cache strategy typo and a generic unrelated issue do not match." - }, - { - "left": "issue:44998", - "right": "issue:45850", - "accept": false, - "reason": "A placeholder title is not enough to merge this with a broad regression complaint." + "reason": "Scheduler kwargs bug is unrelated to the placeholder/unemployment issue." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly grounded, and the 45278 umbrella choice is plausible, but I can\u2019t accept the output because one soft-edge verdict is too strong: 42913 \u2194 43066 is marked true even though the packet only gives titles showing two different tokenizer-v5 regressions (general behavior difference vs wrong decoder type), which is not enough to confirm the same bug. That makes the overall judgment insufficiently conservative." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -15086,8 +15005,8 @@ "review_unit_count": 47, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-400", "nodes": [ @@ -15102,13 +15021,19 @@ "issue:43572", "issue:43618", "issue:44230", + "issue:44492", "issue:44623", "issue:44861", "issue:44998", + "issue:45290", "issue:45310", - "issue:45341" + "issue:45341", + "issue:45850" ], "soft_pairs": [ + "issue:44861|issue:45290", + "issue:44492|issue:44998", + "issue:44998|issue:45850", "issue:43618|issue:44998", "issue:44998|issue:45341", "issue:43525|issue:44861", @@ -15122,106 +15047,124 @@ "issue:31515|issue:33453", "issue:43298|issue:44230" ], - "prepared_review_unit_hash": "66805bcb89922cc3f630a9b9e514b28323c37608cbaaf637c2a0064f1591a789", + "prepared_review_unit_hash": "111f8039cce9a3727b6939c68c609b23c223a5569a22976c763bf66425840a97", "component_budget": { "node_count": 400, "item_count": 400, - "soft_pair_count": 1500, - "serialized_chars": 773310, - "estimated_input_tokens": 193328, - "estimated_eval_tokens": 386912 + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 }, "budget": { - "node_count": 16, - "item_count": 16, - "soft_pair_count": 12, - "serialized_chars": 12531, - "estimated_input_tokens": 3133, - "estimated_eval_tokens": 6522 + "node_count": 19, + "item_count": 19, + "soft_pair_count": 15, + "serialized_chars": 15089, + "estimated_input_tokens": 3773, + "estimated_eval_tokens": 7802 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items form a heterogeneous set of unrelated bug reports; the soft-similarity links are mostly superficial and I would not merge any of the paired issues as duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "issue:31515 is the most connected node in the set, but only by weak, broad loading-related similarity; it is the closest thing to a hub, not a true duplicate center.", + "summary": "The cluster is heterogeneous: it mixes unrelated regressions, a docs typo, a model-config attribute bug, and several loading/runtime issues. None of the soft pairs look like the same underlying bug, so I would not merge any of them as duplicates.", + "confidence": 0.19, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:31515 has the strongest cluster connectivity and the highest apparent discussion signal, so it is the best representative issue here despite the cluster being mixed.", + "best_issue_reason": "Issue 45310 is the most concrete and actionable regression report in the set (a specific from_pretrained failure on Qwen3.5 MoE), but the surrounding issues are too diverse for a true canonical duplicate.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Different failures: tied-weight key handling vs chat template parsing with tool calls. No shared code-path problem." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "A cache-strategy typo is unrelated to an unemployment report." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "Generic broad complaint about many models breaking is not the same as an unemployment issue." + }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "Completely unrelated: a CLIPOutput attention regression versus an off-topic issue with no overlapping code path." + "reason": "CLIP attentions assignment regression is unrelated to the unemployment report." }, { "left": "issue:44998", "right": "issue:45341", "accept": false, - "reason": "No shared bug or subsystem; one is off-topic, the other is a testing_utils bug." + "reason": "Unrelated subjects: unemployment vs a testing_utils bug." }, { "left": "issue:43525", "right": "issue:44861", "accept": false, - "reason": "Both are AttributeError-style failures, but they involve different attributes and different code paths in config vs tied-weight handling." + "reason": "Llama4Config pad_token_id absence and tied-weight key crash are different model/config issues." }, { "left": "issue:43572", "right": "issue:44861", "accept": false, - "reason": "Different regressions: missing pad_token_idx in a config update versus a tied-weights helper crash." + "reason": "StableLmConfig pad_token_idx regression is not the same bug as _get_tied_weight_keys crashing." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Unrelated failures: slow from_pretrained checkpoint loading versus an ImportError in configuration_utils." + "reason": "Slow from_pretrained checkpoint loading and an ImportError for layer_type_validation are unrelated." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "Same broad model/processor area, but different concrete problems: fp8 inference support versus processor.save_pretrained missing files." + "reason": "FP8 Qwen/VL support request is different from processor.save_pretrained missing files." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "Different concerns entirely: MLFlow job status reporting versus checkpoint loading performance." + "reason": "MLFlow job-status reporting and checkpoint loading performance are unrelated." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "No shared underlying bug; checkpoint loading slowness is distinct from tensor-parallel training behavior." + "reason": "Checkpoint loading slowness is not the same as a tensor-parallel training bug." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Loading behavior versus documentation confusion; not the same bug." + "reason": "Loading speed regression and documentation confusion are different classes of issue." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "Related model family, but one is fp8 inference support and the other is a from_pretrained error; different fixes and code paths." + "reason": "Both mention Qwen/from_pretrained-like contexts, but one is FP8 support and the other is a specific loading error; not the same concrete bug." }, { "left": "issue:31515", "right": "issue:33453", "accept": false, - "reason": "Both involve loading, but tokenizer-loading regression is a different failure mode from slow checkpoint loading." + "reason": "Tokenizer loading regression is distinct from slow checkpoint loading." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Different feature gaps for different modalities/models: continuous batching audio support versus fp8 Qwen3 support." + "reason": "Continuous batching audio-model support and FP8 Qwen/VL support are different feature gaps." } ] }, @@ -15240,8 +15183,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-13244-3", "nodes": [ @@ -15253,12 +15196,12 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "1aa040c3b034825a66c7c039b25e694c6c05d62ab1d8f49e39cbbac1c12b70da", + "prepared_review_unit_hash": "f4e96e0d13b14c57a4991de17e4204714c51358c4c93d0cefc666cfd89a07c13", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2477, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -15266,7 +15209,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2477, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -15275,24 +15218,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The three issues appear to be separate: one about Tapas tokenization differing from TensorFlow code, one about a misleading auth/download error for private models, and one about failing to load bert-base-uncased. No duplicate pair is strong enough to merge.", - "confidence": 0.83, - "canonical_issue_reason": "Issue #16618 is the broadest and most user-facing of the set, describing a general model-loading failure rather than a narrow tokenization or auth-message concern.", + "summary": "The cluster only has loose topical similarity around model/token handling. Issue 13244 is about tokenization mismatch, 15980 is about an auth/download error message, and 16618 is a model loading failure. They do not look like duplicates of the same underlying bug.", + "confidence": 0.86, + "canonical_issue_reason": "issue:16618 is the most concrete user-facing runtime failure in the cluster and the broadest model-loading problem; the others are narrower and about different code paths.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #16618 is the best representative because it targets a concrete, commonly encountered load failure and is more central than the other two, which are clearly about different subsystems.", + "best_issue_reason": "issue:16618 is the best anchor for the cluster because it is the clearest concrete bug report, but it still does not subsume the other two issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tapas tokenization mismatch and bert-base-uncased loading failure are different bugs in different code paths; no concrete shared failure mode is evident." + "reason": "Tokenization output differing from TensorFlow code is a different bug class from failing to load a model; the symptoms and likely code paths do not match." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "A private-model login/download error is an authentication/reporting issue, while bert-base-uncased failing to load is a separate model-loading problem; they are not the same underlying bug." + "reason": "Bad error messaging for unauthenticated private-model download is not the same as a generic 'can't load bert-base-uncased' failure; related area, but not the same underlying defect." } ] }, @@ -15311,8 +15254,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-26350-2", "nodes": [ @@ -15322,7 +15265,7 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "27466125b55657fb7b68b68343bbca9345df564a1dda1878de8f6fdd661017e5", + "prepared_review_unit_hash": "3b7ac85596561b41adac32f95e86a3080e054d8dcc787e3bde3d11856dfe6f62", "component_budget": { "node_count": 2, "item_count": 2, @@ -15344,18 +15287,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are both broad community-contribution feature requests, but they target different capabilities: Flash Attention 2 support vs GGUF support. They are related only at a very high level and are not duplicates.", - "confidence": 0.97, - "canonical_issue_reason": "No canonical issue: the two issues request different features for different integration paths, so they should remain separate.", + "summary": "These are two separate community-contribution feature requests: one for Flash Attention 2 architecture support and one for GGUF architecture support. They share a broad theme but not the same underlying bug/change.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the requests target different features and different code paths.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a duplicate of the other; each is the best tracker for its own specific feature request.", + "best_issue_reason": "Neither issue is a duplicate of the other, so there is no single canonical issue for the pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Both are community-contribution architecture-support requests, but one is for Flash Attention 2 and the other is for GGUF. They do not describe the same underlying bug or change." + "reason": "Different requested capabilities (Flash Attention 2 vs GGUF support); same broad subsystem, but not the same underlying change or bug." } ] }, @@ -15374,8 +15317,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-27088-4", "nodes": [ @@ -15392,70 +15335,70 @@ "issue:43834|issue:44365", "issue:44365|issue:44370" ], - "prepared_review_unit_hash": "5fee2463b71fb1b9bbb39c09840c97b1be268b66f0344b9a500348b1c9838b22", + "prepared_review_unit_hash": "1dff98e9c7c67da2e0c92448858e5db7da25cc236f3d9605926697bc785fafb8", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4455, - "estimated_input_tokens": 1114, - "estimated_eval_tokens": 2484 + "serialized_chars": 4452, + "estimated_input_tokens": 1113, + "estimated_eval_tokens": 2482 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4455, - "estimated_input_tokens": 1114, - "estimated_eval_tokens": 2484 + "serialized_chars": 4452, + "estimated_input_tokens": 1113, + "estimated_eval_tokens": 2482 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are highly similar i18n translation request issues, but they do not establish the same concrete language-specific task. The Turkish issue is the only concrete one; the others look like generic template/placeholder variants, so I would not collapse them as duplicates.", - "confidence": 0.61, - "canonical_issue_reason": "Issue 27088 is the earliest concrete request and names a specific language (Turkish). The other issues are generic placeholder titles without a specific language target, so they are less suitable as the canonical representative.", + "summary": "These look like translation-request issues, but only one is concrete (Turkish). The others are generic template-style placeholders and may represent different language requests, so I would not collapse them as duplicates.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 27088 is the most concrete and specific: it names Turkish explicitly and looks like a real translation request rather than a template placeholder.", "canonical_pr_reason": null, - "best_issue_reason": "27088 is the best issue to keep as the cluster representative because it is specific, earliest, and most actionable.", + "best_issue_reason": "Issue 27088 is the strongest representative because it is specific, actionable, and not just a generic language-template title.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Similar template text, but not enough evidence they request the same language translation; this looks like a placeholder variant rather than the same concrete issue." + "reason": "Turkish-specific request vs a generic language-template issue; the similarity appears to come from the shared template, not the same underlying request." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "The bodies may be near-identical, but the issue titles do not identify the same language-specific request, so I would not merge them." + "reason": "Turkish-specific request vs a generic language-template issue; no evidence they target the same language or same translation task." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Turkish is a concrete translation request; the other issue is only a generic placeholder, so they are not clearly the same underlying task." + "reason": "Turkish-specific request vs a generic language-template issue; similar wording alone is not enough to treat them as the same issue." }, { "left": "issue:43834", "right": "issue:44370", "accept": false, - "reason": "Both are generic placeholders, but there is no concrete language target to confirm the same underlying issue." + "reason": "Both appear to be template-style translation requests, but there is no concrete evidence they are the same language or the same underlying task." }, { "left": "issue:43834", "right": "issue:44365", "accept": false, - "reason": "Near-identical template text alone is not enough to prove a duplicate when no specific language request is stated." + "reason": "Identical placeholder-style titles are not enough to merge; they may correspond to different language-specific requests." }, { "left": "issue:44365", "right": "issue:44370", "accept": false, - "reason": "These appear to be template/placeholder variants; without a named language they are not safely the same concrete issue." + "reason": "Both are generic translation-request templates, but no explicit language target or shared concrete bug/change is shown." } ] }, @@ -15474,8 +15417,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-36817-15", "nodes": [ @@ -15532,136 +15475,136 @@ "issue:37015|issue:45027", "issue:43837|issue:45177" ], - "prepared_review_unit_hash": "6b8892407e6c96ea876e0f5c1121d7cdbd70ffe44c6b94dd537aca15e7d3142f", + "prepared_review_unit_hash": "ea29828d0bdc27ecc969455f3ab60d3576f0860d8e95d5078852e49cbba68b75", "component_budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 35, - "serialized_chars": 20707, - "estimated_input_tokens": 5177, - "estimated_eval_tokens": 10610 + "serialized_chars": 20712, + "estimated_input_tokens": 5178, + "estimated_eval_tokens": 10612 }, "budget": { "node_count": 15, "item_count": 15, "soft_pair_count": 35, - "serialized_chars": 20707, - "estimated_input_tokens": 5177, - "estimated_eval_tokens": 10610 + "serialized_chars": 20712, + "estimated_input_tokens": 5178, + "estimated_eval_tokens": 10612 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a set of distinct model-request issues, not duplicates; the soft matches are driven by shared wording like \u201cAdd model\u201d rather than the same underlying change.", - "confidence": 0.94, - "canonical_issue_reason": null, + "summary": "This cluster is a loose theme cluster of model-add/support requests, not true duplicates. The issues name different models or different modalities, so they should stay separate; at most, a couple are nearby subfamily requests.", + "confidence": 0.84, + "canonical_issue_reason": "If one issue must act as the anchor, issue 43837 is the best fit because it sits in the most connected subcluster and has the strongest overlap with a nearby related request (43671), but the set is still not duplicate enough to collapse into one issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43837 is the clearest representative of the cluster\u2019s higher-similarity audio/model-request subgroup, but it still does not subsume the other requests.", + "best_issue_reason": "Issue 43671 is the best representative of the cluster because it is a clear, specific model-support request with substantial discussion and belongs to the densest nearby subcluster; however, it still represents a distinct request from the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Different model requests: EuroBert config support vs DEIMv2 architecture support." + "reason": "Different model requests (EuroBert vs DEIMv2); same broad pattern, not the same bug/change." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert config support and Qwen3-TTS support are unrelated model additions." + "reason": "Different requested models and different modality area; not a duplicate." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert and RF-DETR are different model families with different integration targets." + "reason": "Both are model-add requests, but for unrelated models and code paths." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are separate model additions." + "reason": "RF-DETR and DEIMv2 are distinct model integrations." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "Object-detection RF-DETR vs audio model LFM2.5 Audio 1.5B are unrelated." + "reason": "Detection model request vs audio model request; not the same change." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "RF-DETR and the weight-sparse transformer request do not describe the same code path or model." + "reason": "Different model families and integration work; no single concrete fix." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR and Qwen3-ASR are different models in different domains." + "reason": "Different requested models; same support-request theme only." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR and Qwen3-TTS are distinct model support requests." + "reason": "RF-DETR vs Qwen3-TTS are unrelated model additions." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR are unrelated model integrations." + "reason": "NeoBERT and Qwen3-ASR are separate model support requests." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT request vs OpenAI weight-sparse transformer request are different changes." + "reason": "Different models with different implementation paths." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT and LFM2.5 Audio 1.5B are not the same model or fix." + "reason": "NeoBERT and LFM2.5 Audio are unrelated requests." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are separate model families and requests." + "reason": "Different models; only the generic request template matches." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are different model requests, not one duplicate." + "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 model additions." + "reason": "MiniCPM3 architecture request vs Qwen3-ASR support; different artifacts." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 and the weight-sparse transformer request do not target the same underlying model change." + "reason": "Different model families and likely different files/code paths." }, { "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 and Voxtral TTS are different model integrations." + "reason": "MiniCPM3 architecture vs Voxtral TTS support; not the same change." }, { "left": "issue:41211", @@ -15673,103 +15616,103 @@ "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are separate model additions." + "reason": "ModernVBERT and Qwen3-ASR are different models." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are different model requests." + "reason": "ModernVBERT models vs Qwen3-TTS support; not a duplicate." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT vs generic BERT-like models with RoPE is related in theme only, not the same concrete change." + "reason": "ModernVBERT and BERT-like models with RoPE are different requests; too broad to merge." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like models with RoPE and Voxtral TTS do not refer to the same model or code path." + "reason": "A generic BERT-like/RoPE request is not the same as Voxtral-4B-TTS support." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR are related by naming only; they are separate speech tasks and integrations." + "reason": "Same Qwen3 family, but TTS vs ASR are different products and code paths." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Qwen3-TTS and Voxtral-4B-TTS-2603 are different TTS model requests." + "reason": "Both are speech-related, but they request different model support." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated modality/model requests." + "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated model additions." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "Qwen3-TTS and the weight-sparse transformer request are not the same change." + "reason": "Different model families and different integration work." }, { "left": "issue:43671", "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are different audio model additions." + "reason": "Qwen3-TTS and LFM2.5 Audio are distinct audio-model requests." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio 1.5B are separate speech/audio model requests." + "reason": "Qwen3-ASR and LFM2.5 Audio are not the same model or fix." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR and EUPE are different model requests with different purposes." + "reason": "Qwen3-ASR support and EUPE are unrelated model requests." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR support and the weight-sparse transformer request are unrelated." + "reason": "Qwen3-ASR and the weight-sparse transformer are distinct requests." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio 1.5B and SAM3-LiteText are different model additions." + "reason": "Audio model support vs SAM3-LiteText; different model integrations." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "OpenAI weight-sparse transformer and EUPE are different model requests." + "reason": "Weight-sparse transformer and EUPE are separate model requests." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral-4B-TTS-2603 are distinct speech-model integrations." + "reason": "Both are speech-related, but ASR and TTS support are different concrete changes." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral-4B-TTS-2603 are unrelated model requests." + "reason": "NeoBERT and Voxtral-4B-TTS are unrelated model additions." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR and DeepSeek-OCR2 are different modality/model support requests." + "reason": "Qwen3-ASR and DeepSeek-OCR2 do not target the same model or code path." } ] }, @@ -15788,8 +15731,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-37934-6", "nodes": [ @@ -15810,82 +15753,82 @@ "issue:43668|issue:44031", "issue:43668|issue:43813" ], - "prepared_review_unit_hash": "2c4ffa977cff7e6ded4b6666086f18b2c61142db3e0267246fe56522be837d48", + "prepared_review_unit_hash": "6f142b0557b4ef80729795269016402f28030efc2717b63aa38ccea90564d478", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 8, - "serialized_chars": 6288, - "estimated_input_tokens": 1572, - "estimated_eval_tokens": 3400 + "serialized_chars": 6291, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 8, - "serialized_chars": 6288, - "estimated_input_tokens": 1572, - "estimated_eval_tokens": 3400 + "serialized_chars": 6291, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These issues are not duplicates of one another; they span different models, configs, utilities, and symptoms, so there is no single canonical cluster representative.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "These issues are only loosely text-similar and describe different bugs across unrelated areas (model norms, config typing, loss kwargs, a typo, tokenizer warnings). I do not see a true duplicate cluster here.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue: the items cover unrelated problem types and code paths, so none is a good duplicate anchor.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "No best issue for deduplication because the set is heterogeneous rather than one underlying bug.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Different components and bug types: Llama4 normalization behavior vs ModernBERT config type hint. No shared code path or symptom." + "reason": "Both mention normalization/config names, but one questions an architecture choice in Llama4 while the other is a type-hint issue in ModernBERTConfig; not the same bug." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Mask2Former ignore_value usage in model setup is unrelated to kwargs not being forwarded into a loss function." + "reason": "Mask2Former ignore_value handling and loss kwargs plumbing are different parts of the training stack and different failure modes." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Different subsystems and failures: Mask2Former loss handling vs tokenizer regex warnings after a version change." + "reason": "Unrelated topics: a model loss parameter not used versus tokenizer regex warning regression." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Mask2Former training issue vs a typo in a quantization attribute name; these are unrelated bugs." + "reason": "A Mask2Former loss-parameter issue is unrelated to a typo in a quantization-operation attribute name." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Model-specific training parameter usage vs a config annotation mistake in a different model." + "reason": "Different models and different bugs: ignore_value usage versus an incorrect type hint." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Loss-kwargs propagation bug and ModernBERT type-hint issue do not share the same concrete code path." + "reason": "Loss-kwargs propagation and ModernBERTConfig type annotation are separate issues with no shared code path." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "A config type-hint issue is not the same as tokenizer regex warning behavior; no common fix." + "reason": "A config type hint bug does not match a tokenizer-regex warning regression." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "Different modules and bug categories: config typing vs a misspelled attribute reference." + "reason": "ModernBERTConfig typing and an attribute-name typo are distinct problems in different code areas." } ] }, @@ -15904,8 +15847,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-41084-2", "nodes": [ @@ -15915,12 +15858,12 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "101da398ff831d64b3840e966df55e820063e7ae8add46eb4f400cfa46dd7f6b", + "prepared_review_unit_hash": "f2f820abf20784a482a6536760011db0bb596b7a6121c03cc6d888fbfaf5586e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, + "serialized_chars": 1478, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -15928,7 +15871,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1480, + "serialized_chars": 1478, "estimated_input_tokens": 370, "estimated_eval_tokens": 996 }, @@ -15937,18 +15880,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues do not look like duplicates: one is about block decoding behavior, the other is a feature/support request for DeepSeek v3.2. They appear to be separate topics with no shared code-path evidence.", - "confidence": 0.9, - "canonical_issue_reason": null, + "summary": "The two issues are not duplicates: one is a vague feature request for \"Set Block Decoding,\" while the other is a specific request for \"DeepSeek v3.2 support.\" They appear to be unrelated product/runtime feature requests rather than the same underlying problem.", + "confidence": 0.92, + "canonical_issue_reason": "Issue #43037 is the better canonical anchor because it is more specific and actionable; #41084 is too vague to serve as a stable representative for the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "\"DeepSeek v3.2 support\" is the more specific and actionable issue title; \"Set Block Decoding\" is too vague to serve as a strong canonical reference.", + "best_issue_reason": "#43037 is the best standalone issue candidate due to its concrete scope and clearer target. #41084 is too underspecified to be the primary reference issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "Set Block Decoding", - "right": "DeepSeek v3.2 support", + "left": "issue:41084", + "right": "issue:43037", "accept": false, - "reason": "Different underlying concerns: decoding configuration versus model-version support. No shared filenames, targets, or concrete bug path indicating the same issue." + "reason": "The titles and available metadata point to different requests: block decoding behavior versus DeepSeek v3.2 model support. No shared files, targets, or issue links suggest the same underlying change." } ] }, @@ -15967,8 +15910,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43598-2", "nodes": [ @@ -15978,7 +15921,7 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "087a41c13a73690fb9c486d49d44c34f4a558d0d66eed3bce037e4ff6c457aca", + "prepared_review_unit_hash": "ea4e7f13aec445d58bcdb636eb9f774f112fcc06e0aa978d7f87c8862aa99c46", "component_budget": { "node_count": 2, "item_count": 2, @@ -16000,18 +15943,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues both mention custom trainers, but they target different Trainer behaviors: one is about `remove_unused_column` customizability, the other about estimating FLOPs correctly. They are related at a broad subsystem level, but not the same underlying bug.", - "confidence": 0.93, + "summary": "These two issues are about different trainer behaviors: one concerns `remove_unused_column` customizability, the other concerns FLOPS estimation for custom trainers. They do not appear to describe the same underlying bug or change.", + "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No strong canonical issue here; if one must be selected, #43598 is the more concrete user-facing Trainer customization report, while #43600 is a separate FLOPs-accounting concern.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Different code-path problems: #43598 is about `remove_unused_column` behavior/customizability, while #43600 is about FLOPs estimation for custom trainers. Shared wording around custom trainers is too broad to treat as duplicates." + "reason": "Different problem domains and code paths: one is about column pruning customization in Trainer, the other about FLOPS accounting for custom trainers. Shared theme is only broad Trainer customization, not a duplicate bug/change." } ] }, @@ -16030,8 +15973,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43684-2", "nodes": [ @@ -16041,12 +15984,12 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "a5a9ffc4bbf514be4713abfa1c78e3765d223a5c466b37673e7a8bf80c119322", + "prepared_review_unit_hash": "c42d55c30f6c06de9da09976bc911fd47df1406b9fad1a8084703601c9c1e033", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -16054,7 +15997,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1611, + "serialized_chars": 1612, "estimated_input_tokens": 403, "estimated_eval_tokens": 1062 }, @@ -16063,18 +16006,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are both feature requests around Qwen model support, but they target different model families and different AutoModel entry points, so they do not look like duplicates.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical issue: issue #43684 is about registering Qwen3-Omni with AutoModel/AutoModelForConditionalGeneration, while issue #44405 is about AutoModelForSequenceClassification support for Qwen3.5.", + "summary": "Both issues are about extending Hugging Face auto-class support for Qwen-family models, but they request different changes: Qwen3-Omni registration in core auto classes vs. Qwen3.5 sequence-classification support. They are related feature requests, not the same underlying bug/change.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43684 is the broader auto-registration request and best represents the cluster\u2019s general theme, but it is still not the same as the Qwen3.5 sequence-classification request.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #43684 is the broader, more central request about model registration in the AutoModel stack, so it is the better representative issue for this cluster.", + "best_issue_reason": "issue:43684 is the better cluster representative because it targets core AutoModel/AutoModelForConditionalGeneration registration rather than a narrower task-specific head.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different model families and different tasks/APIs: Qwen3-Omni registration vs Qwen3.5 sequence classification support. They do not appear to be the same underlying bug or change." + "reason": "Different model variants and different auto-class targets; related in theme, but not the same concrete issue or change." } ] }, @@ -16093,8 +16036,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43835-4", "nodes": [ @@ -16111,76 +16054,79 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "f3da97c7bb0059b08b9d35b4623b75f46921a19a9e66f508c611dc6d0b82966e", + "prepared_review_unit_hash": "46073e5cddb5c8499c487ae96083e3eca4e454d69aaeae59b362c4f3cfc03ee8", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4265, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4258, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4265, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4258, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All four issues are extremely low-information and only weakly similar by title. There is not enough evidence to treat any pair as the same underlying report, so no soft duplicates are accepted.", - "confidence": 0.22, - "canonical_issue_reason": "If a single issue must be picked, issue #43835 is the least generic and earliest, but it still lacks substantive detail and is not a strong canonical source.", + "summary": "#43845 is the best canonical issue among the three near-identical \"huggingface \" reports because it was opened first. #43835 looks unrelated. The similarity between the huggingface issues is too shallow to confidently mark any pair as duplicates without body details or explicit links.", + "confidence": 0.78, + "canonical_issue_reason": "#43845 is the earliest of the three near-identical \"huggingface \" issues, so it is the best representative of that cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #43835 is slightly preferable only because its title is more distinct than the repeated 'huggingface' entries and it was created first; otherwise none is clearly suitable.", + "best_issue_reason": "#43845 is the strongest issue candidate because it is the first report in the placeholder-title cluster and is the most natural canonical representative.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Shared only vague similarity; no body details or explicit targets to show the same underlying bug." + "reason": "Different titles and no concrete shared bug context; the evidence is only weak text similarity." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Both are low-signal, but there is no concrete evidence they describe the same problem." + "reason": "Different titles and no shared details point to unrelated reports rather than the same issue." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Title overlap is too weak to merge as duplicates without supporting description or references." + "reason": "\"Fett\" does not plausibly describe the same underlying problem as the huggingface issues." }, { "left": "issue:43845", "right": "issue:43846", "accept": false, - "reason": "Near-identical titles alone are insufficient; no distinguishing evidence of the same issue." + "reason": "Same placeholder title, but only soft similarity is available and deterministic acceptance is false, so this is not enough to confirm a duplicate." }, { "left": "issue:43845", "right": "issue:43859", "accept": false, - "reason": "Possible textual duplication, but no concrete report content confirms a shared underlying bug." + "reason": "The titles are similar, but there are no body details or explicit links tying them to the same underlying bug." }, { "left": "issue:43846", "right": "issue:43859", "accept": false, - "reason": "Insufficient evidence beyond generic title similarity to treat these as duplicates." + "reason": "These look superficially alike, but the available evidence is too weak to treat them as duplicates with confidence." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The canonical issue choice is reasonably grounded: #43845 is the earliest of the three issues with the same title. However, the soft-edge verdicts are too strong for the packet: all pairings are supported only by soft_similarity, and the packet\u2019s deterministic_accept is false for every pair. With no issue-body details or explicit links, it is not conservative to mark the #43845/#43846/#43859 pairs as duplicates with high confidence. Rejecting the #43835 comparisons is fine, but the duplicate conclusions among the huggingface issues overstate the evidence." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -16193,8 +16139,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-44371-5", "nodes": [ @@ -16215,88 +16161,88 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "2085834173fd0497841a5147f4bac426e759c680bcbd3933d984c1c965ea2739", + "prepared_review_unit_hash": "5240be50cddc1c18171a3108f04aac3f19fa353feaeeb470bc99c61a7ce9c13e", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5785, - "estimated_input_tokens": 1447, - "estimated_eval_tokens": 3150 + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5785, - "estimated_input_tokens": 1447, - "estimated_eval_tokens": 3150 + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These items do not look like duplicates of one underlying issue. Four are spam-like or vague titles, and the remaining one is only a generic documentation suggestion with no concrete shared bug/change signal. The soft-similarity links are not enough to justify merging them.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "These are not clear duplicates of one bug/feature; they look like unrelated low-signal/spam issues with only superficial textual similarity. No soft pair should be merged.", + "confidence": 0.87, + "canonical_issue_reason": "issue:44418 is the least spam-like and most issue-shaped title in the set, so it is the best forced representative if one must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44418 is the only non-spam, plausibly legitimate report in the set, but it is still too vague to serve as a true duplicate anchor.", + "best_issue_reason": "issue:44418 is the most plausible standalone issue title here; the others are generic spam placeholders or unrelated text.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are spam-like '' titles with no substantive description or shared concrete problem." + "reason": "Both are just '' with no concrete shared bug or change; no evidence they describe the same underlying issue." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both are spam-like '' titles; no evidence they describe the same bug or request." + "reason": "Both are spam placeholders and provide no shared problem description beyond the title text." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam-like and the other is a vague documentation suggestion; they do not share a concrete underlying issue." + "reason": "One is spam and the other is a generic documentation suggestion; not the same underlying issue." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "A spam-like issue and an unrelated title about 'Google LLC'; no sign of the same underlying report." + "reason": "'' and 'Check out \"Google LLC\"' are unrelated titles with no concrete bug overlap." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "A generic documentation suggestion and a spam placeholder are not the same issue." + "reason": "Documentation enhancement request does not match a spam placeholder issue; no shared problem statement." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "A generic documentation suggestion and a spam placeholder are not the same issue." + "reason": "Generic documentation request and spam placeholder are unrelated." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are spam-like placeholders, but there is no concrete bug/change detail tying them to one underlying issue." + "reason": "Same spam marker, but that alone is not enough to indicate the same underlying bug or change." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus unrelated 'Google LLC' title; no shared concrete problem." + "reason": "Spam placeholder and unrelated Google LLC title do not point to the same issue." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder versus unrelated 'Google LLC' title; no shared concrete problem." + "reason": "Spam placeholder and unrelated Google LLC title do not describe the same bug or request." } ] }, @@ -16315,8 +16261,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45145-2", "nodes": [ @@ -16326,12 +16272,12 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "12ed42f8e2c7599c4b95808bdd4ea21fbfe89314939b3ec2620173d10f42c8cd", + "prepared_review_unit_hash": "44f6144620f78e303387c8dab336fc1e19178192af2e148f87351790478c33c9", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1646, + "serialized_chars": 1645, "estimated_input_tokens": 412, "estimated_eval_tokens": 1080 }, @@ -16339,7 +16285,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1646, + "serialized_chars": 1645, "estimated_input_tokens": 412, "estimated_eval_tokens": 1080 }, @@ -16348,28 +16294,28 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Likely duplicate closed issues with identical titles and similar body lengths, but the evidence available here is limited to metadata and a soft-similarity signal. The cluster should be treated as a probable duplicate pair, not a proven near-identical match.", - "confidence": 0.82, - "canonical_issue_reason": "Issue #45145 is the better canonical record because it was created first and has the only inbound reference, making it the more established issue page.", + "summary": "The two issues are highly similar and may be duplicates: they share the exact title, have very close body lengths, and there are no distinguishing links, targets, or filenames. However, the packet does not prove they are exact duplicates, so the safer conclusion is likely-duplicate rather than confirmed duplicate. #45145 is the better canonical record because it was created first and has an inbound reference.", + "confidence": 0.83, + "canonical_issue_reason": "#45145 is the best canonical issue because it is earlier and has an inbound reference. The evidence supports strong similarity with #45151, but not a definitive exact-duplicate determination.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #45145 is the best representative issue for this cluster for the same reasons: earliest creation and slightly stronger linkage history.", + "best_issue_reason": "#45145 is the best representative issue for the cluster: it is the earlier report and has slightly more established history via the inbound reference, while #45151 does not add distinct scope.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", - "accept": true, - "reason": "Same title and a soft-similarity signal indicate they likely describe the same issue, and there is no metadata suggesting different scope. The evidence is not enough to claim identical bodies, but it is sufficient for a duplicate cluster." + "accept": false, + "reason": "The pair looks highly similar, but the evidence is only soft similarity plus matching titles. With no explicit links, targets, or content proof, this is not enough to confidently accept them as the same underlying issue." } ] }, "evaluator_result": { "accept": false, - "feedback": "The comparison is not supported as strongly as stated. The packet only shows identical titles, similar body lengths, and a soft_similarity signal; it does not provide the issue bodies or evidence that the descriptions/content are the same. The cluster may still be a plausible duplicate pair, but the summary and soft-edge rationale overstate the evidence by claiming near-perfect duplication and identical issue descriptions." + "feedback": "The canonical choice is plausible, but the writeup overstates the evidence. The packet supports that the two issues have the same title and very similar metadata/body lengths, with no explicit links or targets, but it does not establish that they are exact duplicates or that the issue content is the same. The soft pair is also only a similarity signal and the deterministic decision is false, so the duplicate verdict and 0.99 confidence are too strong. A more conservative summary would say they appear highly similar / likely duplicates, with #45145 as the better canonical record because it is earlier and has an inbound reference." }, "evaluator_used": true, "retried": true, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -16435,8 +16381,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45672-2", "nodes": [ @@ -16446,7 +16392,7 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "953b1ad1d73f6a41f7ad14a78ae94b6f6d2811e4b28fc8041955f215f481bb21", + "prepared_review_unit_hash": "65b21d096078d27c4cceb409f978dd81a76bcc6de62d377dfa17d102c8c99c21", "component_budget": { "node_count": 2, "item_count": 2, @@ -16468,18 +16414,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate 4-bit inference bugs. 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 triggers an AssertionError. They share only a broad quantization context, not the same underlying failure or fix.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "These two issues both mention 4-bit inference, but they describe different failure modes and likely different code paths: one is a Gemma4/audio-module uint8 dtype error, the other is a BitsAndBytesConfig skip-module regression that removes the default lm_head exclusion. I would keep them separate and not merge the cluster.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45674 is the better cluster representative because it describes a concrete, broadly applicable BitsAndBytesConfig regression with a clear expected/actual behavior and an obvious configuration-level fix path.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45674 is the clearer representative bug report because it names the specific BitsAndBytesConfig regression and concrete AssertionError, but it does not subsume the Gemma4/audio torch.finfo failure in 45672.", + "best_issue_reason": "#45674 is the strongest issue to anchor on: it is specific, reproducible, and appears to describe the main actionable bug in the cluster. #45672 is a narrower model-specific dtype crash in audio modules and does not match the same root cause.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Different bug paths and fixes: 45672 is a Gemma4/audio uint8 torch.finfo TypeError, while 45674 is a BitsAndBytesConfig llm_int8_skip_modules regression that removes lm_head exclusion and causes an AssertionError. Shared 4-bit inference context is not enough to merge them." + "reason": "Both are 4-bit inference bugs, but the failure mechanisms differ: #45672 is a Gemma4/audio uint8 torch.finfo TypeError, while #45674 is a BitsAndBytesConfig skip-module bug causing lm_head exclusion to be cleared and an AssertionError. They do not look like the same underlying issue." } ] }, @@ -16498,8 +16444,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45762-2", "nodes": [ @@ -16509,40 +16455,40 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "9a7fe8996ed6b865fd628879d7fea85aecc9723603f0e36b0ff6cd2daf40b83b", + "prepared_review_unit_hash": "7d1384808e2e81500921827e3a7a187a03629af2c2f7ed4f132e07478de6c054", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1476, - "estimated_input_tokens": 369, - "estimated_eval_tokens": 994 + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1476, - "estimated_input_tokens": 369, - "estimated_eval_tokens": 994 + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two closed issues appear unrelated: they are just distinct person-name/date strings with no body, code references, or shared targets. The soft similarity signal is not enough to treat them as duplicates.", - "confidence": 0.98, - "canonical_issue_reason": null, + "summary": "Both items look like unrelated low-information issues/spam entries (personal names plus birth dates), not duplicate bug reports. There is no meaningful canonical issue or PR here.", + "confidence": 0.97, + "canonical_issue_reason": "None: neither issue contains a bug report, repro, or substantive discussion to serve as a canonical duplicate target.", "canonical_pr_reason": null, - "best_issue_reason": "Neither issue is a strong canonical candidate because neither contains actionable bug/report content, and there is no evidence they describe the same problem.", + "best_issue_reason": "Neither issue is suitable as a global best issue; both are empty placeholder-like entries with no actionable content.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "The titles differ in both name and date, and there is no body, linkage, or technical context indicating the same underlying issue." + "reason": "The titles are just different personal names and dates; there is no shared bug, feature, or code-path problem to indicate duplication." } ] }, @@ -16561,8 +16507,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45764-2", "nodes": [ @@ -16572,7 +16518,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "8c3313ac973c53141626c5b159b2de4496c21e0e80410f415b10781b0fb42479", + "prepared_review_unit_hash": "110af76adcf80338b6b3aa8304f9685c59f559825f900f0c9c81faedbffdeb5a", "component_budget": { "node_count": 2, "item_count": 2, @@ -16594,18 +16540,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two closed issues with only person-name/date titles and no body or technical detail. They do not appear to describe the same bug or change; the similarity is just the shared surname.", - "confidence": 0.96, - "canonical_issue_reason": "Neither issue has actionable content, but issue #45764 is slightly more concise and can serve as the representative placeholder for this cluster.", + "summary": "The two issues are just separate name-and-date titles with no body, links, or shared context; there is no evidence they describe the same bug or request.", + "confidence": 0.95, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue #45764 is the cleanest representative only because it is as information-poor as the other item and slightly simpler; there is no substantive technical distinction.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "Reject: both are closed, content-free title-only issues, and the overlap is only a shared surname/date pattern rather than a shared underlying bug or request." + "reason": "No substantive overlap beyond both being bare personal-name titles; they do not appear to report the same underlying problem." } ] }, @@ -16624,8 +16570,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45869-5", "nodes": [ @@ -16645,12 +16591,12 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "8b547eecf8e16b75b3449fd44d0b8400f4fc482417fc1331f52ab978b07ba70c", + "prepared_review_unit_hash": "d44579cb77a0837216217ddc5acd11ab1afa90d63ab85500a28997e9dc6d6dee", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5728, + "serialized_chars": 5726, "estimated_input_tokens": 1432, "estimated_eval_tokens": 3120 }, @@ -16658,7 +16604,7 @@ "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5728, + "serialized_chars": 5726, "estimated_input_tokens": 1432, "estimated_eval_tokens": 3120 }, @@ -16667,66 +16613,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two exact-duplicate issue pairs are present: 45870/45871 and 45872/45873. Issue 45869 is only loosely related by theme and is not a duplicate of the others.", - "confidence": 0.93, - "canonical_issue_reason": "45872 is the strongest canonical issue: it is the more specific request, and 45873 is an exact duplicate. The 45870/45871 pair appears to be a separate duplicate issue topic.", + "summary": "Two exact duplicate issue pairs are present: 45870/45871 and 45872/45873. Issue 45869 appears to be a separate dataset/sample request and does not match the others.", + "confidence": 0.97, + "canonical_issue_reason": "45870 is the earliest of the exact duplicate pair 45870/45871 and has the cleanest representative title for that topic.", "canonical_pr_reason": null, - "best_issue_reason": "45872 is the best representative issue overall because it is the clearest, most concrete financial-data/service request and has an exact duplicate counterpart; the other similar titles are separate requests, not the same underlying issue.", + "best_issue_reason": "45870 is the best single issue to keep for its duplicate pair because it is the earliest, clear, and exact match to 45871.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Same general finance theme, but one is a dataset bundle listing and the other is an example request; not the same underlying issue." + "reason": "Different requests: a financial dataset bundle versus a DataForge prediction example." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Dataset bundle vs. example request; related topic only, not a duplicate." + "reason": "Different requests: a financial dataset bundle versus a DataForge prediction example." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Identical title and matching metadata suggest the same example request duplicated twice." + "reason": "Exact duplicate titles and same described example request." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both mention financial time series/DataForge, but they ask for different deliverables: an example versus a scraping/service feature." + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Different requested change; similar finance/DataForge wording is not enough to treat them as the same issue." + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Exact duplicate of the example request, but 45872 is a different feature request." + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "The example request and the scraping/service request are distinct underlying issues." + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Identical title and matching metadata indicate a direct duplicate pair." + "reason": "Exact duplicate titles and same described DataForge scraping-service request." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet. The exact duplicate pairs 45870/45871 and 45872/45873 are supported by identical titles and matching body length, and the non-duplicate judgments for 45869 versus the other issues are conservative. No evidence is overstated." + "feedback": "Grounded overall. The summary conservatively separates 45869 from the two duplicate-looking pairs, and the pair verdicts match the packet: 45870/45871 and 45872/45873 have identical titles and jaccard 1.0, while the cross-topic pairs are reasonably rejected." }, "evaluator_used": true, "retried": false, @@ -16745,7 +16691,7 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-39895-17", + "cluster_id": "cluster-39895-16", "nodes": [ "pull_request:39895", "pull_request:41855", @@ -16762,8 +16708,7 @@ "pull_request:45795", "pull_request:45899", "pull_request:45919", - "pull_request:45922", - "pull_request:45946" + "pull_request:45922" ], "soft_pairs": [ "pull_request:39895|pull_request:43424", @@ -16782,238 +16727,231 @@ "pull_request:43995|pull_request:45213", "pull_request:43995|pull_request:44827", "pull_request:43995|pull_request:45899", - "pull_request:44827|pull_request:45946", "pull_request:44827|pull_request:45899", "pull_request:44827|pull_request:45170", "pull_request:44827|pull_request:45795", + "pull_request:44827|pull_request:45655", "pull_request:45170|pull_request:45899", "pull_request:44827|pull_request:45213", "pull_request:43823|pull_request:45213", "pull_request:45213|pull_request:45899", - "pull_request:44827|pull_request:45655", "pull_request:45795|pull_request:45899", "pull_request:43823|pull_request:45795", - "pull_request:45899|pull_request:45946", + "pull_request:43823|pull_request:45899", "pull_request:44827|pull_request:45922", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "4f6f578a969fcd97bb2572e6ae954b0932385f7c4f9112b43707087e84aaefbb", + "prepared_review_unit_hash": "a33177a03aea6443bb15300d6f1cb27231bb9e303e6daebbc911e61023395bd2", "component_budget": { - "node_count": 17, - "item_count": 17, - "soft_pair_count": 30, - "serialized_chars": 49253, - "estimated_input_tokens": 12314, - "estimated_eval_tokens": 24884 + "node_count": 16, + "item_count": 16, + "soft_pair_count": 29, + "serialized_chars": 46699, + "estimated_input_tokens": 11675, + "estimated_eval_tokens": 23606 }, "budget": { - "node_count": 17, - "item_count": 17, - "soft_pair_count": 30, - "serialized_chars": 49253, - "estimated_input_tokens": 12314, - "estimated_eval_tokens": 24884 + "node_count": 16, + "item_count": 16, + "soft_pair_count": 29, + "serialized_chars": 46699, + "estimated_input_tokens": 11675, + "estimated_eval_tokens": 23606 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is dominated by unrelated Transformers maintenance, docs, and model-addition PRs. The only genuine duplicate-like pair is the two `fixed_cross_entropy` fixes (43251 and 43254), which address the same bug in the same code path; 43251 is a reasonable representative, not uniquely canonical.", - "confidence": 0.89, + "summary": "The cluster is mostly heterogeneous: model additions, docs updates, refactors, and unrelated fixes. Only one true duplicate pair stands out: the two `fixed_cross_entropy` PRs that both implement the same kwargs support in `src/transformers/loss/loss_utils.py` for issue 43240.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43251 is a reasonable representative of the shared `fixed_cross_entropy` bugfix because it explicitly targets issue 43240 and forwards kwargs into `nn.functional.cross_entropy`; PR 43254 makes the same underlying change in an equivalent form, so 43251 is not uniquely canonical.", + "canonical_pr_reason": "PR 43251 is the clearest canonical representative of the only real duplicate pair: it targets issue 43240 and makes the concrete `fixed_cross_entropy` kwargs change in `loss_utils.py`.", "best_issue_reason": null, - "best_pr_reason": "PR 43251 is a solid representative for the cluster\u2019s only true duplicate pair since it fixes the concrete `fixed_cross_entropy` kwargs bug tied to issue 43240. That said, 43254 is essentially the same fix, so this is a representative choice rather than a uniquely best one.", + "best_pr_reason": "PR 43251 is the strongest single PR to keep as representative because it directly fixes the linked loss issue with the exact same code-path change as 43254.", "soft_edge_verdicts": [ { "left": "pull_request:39895", "right": "pull_request:43424", "accept": false, - "reason": "VideoPrism model/docs addition vs ExecuTorch dynamic-shape export test; unrelated changes." + "reason": "Unrelated changes: VideoPrism model/docs addition vs ExecuTorch dynamic-shape export test." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Mistral tokenizer API fix vs new MobileLLM model implementation; different code paths." + "reason": "Different artifacts: Mistral tokenizer method additions vs adding a new MobileLLM model." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Tokenizer method addition vs export test; no shared underlying bug or change." + "reason": "Tokenizer API additions vs ExecuTorch export test; no shared bug or change." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor robustness refactor vs new MobileLLM model; unrelated work." + "reason": "Processor robustness fix vs new model implementation; not the same underlying change." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Loss-function bug fix vs new model addition; not the same change." + "reason": "Loss utility kwargs fix vs MobileLLM model add; unrelated." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "`fixed_cross_entropy` kwargs fix vs Mistral4 test/model work; unrelated." + "reason": "Both touch training-related code, but cross-entropy kwargs support and Mistral4 test fixes are different changes." }, { "left": "pull_request:43424", "right": "pull_request:45919", "accept": false, - "reason": "ExecuTorch export test vs Sapiens2 model addition; different artifacts and code paths." + "reason": "ExecuTorch export test and Sapiens2 model addition are unrelated." }, { "left": "pull_request:43424", "right": "pull_request:45899", "accept": false, - "reason": "Dynamic-shape export test vs continuous-batching docs update; not the same bug or change." + "reason": "ExecuTorch exportability test vs docs-only continuous batching update; different code paths." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Export test vs Falcon/Mistral4 refactoring and tests; unrelated." + "reason": "Dynamic-shape export test vs Mistral4 test/model work; no shared fix." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Export test vs new MobileLLM model implementation; unrelated." + "reason": "ExecuTorch export testing and new model scaffolding are unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "MobileLLM model addition vs config-resolution-order fix; different issues." + "reason": "New MobileLLM model vs config-resolution ordering fix; not the same bug." }, { "left": "pull_request:43823", "right": "pull_request:45922", "accept": false, - "reason": "New model addition vs vision-model memory leak fix; unrelated subsystems." + "reason": "New model addition vs lru-cache memory leak fix in vision models; different changes." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM addition vs Falcon output-interface refactor; not the same code-path problem." + "reason": "New model addition vs Falcon output-collection refactor; unrelated." }, { "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "Falcon refactor vs model-creation skill/docs PR; unrelated." + "reason": "Falcon refactor vs model-creation skill docs; not the same artifact or fix." }, { "left": "pull_request:43995", "right": "pull_request:44827", "accept": false, - "reason": "Falcon output-collection refactor vs Mistral4 test/model fixes; different changes." + "reason": "Falcon interface refactor and Mistral4 test fix are different code-path changes." }, { "left": "pull_request:43995", "right": "pull_request:45899", "accept": false, - "reason": "Falcon model refactor vs continuous-batching docs; unrelated." + "reason": "Falcon model refactor vs docs-only decode fast-path note; unrelated." }, { "left": "pull_request:44827", - "right": "pull_request:45946", + "right": "pull_request:45899", "accept": false, - "reason": "Mistral4 test/model fix vs RF-DETR release-org/docs update; unrelated." + "reason": "Mistral4 test/model work vs docs-only continuous batching update; no common bug." }, { "left": "pull_request:44827", - "right": "pull_request:45899", + "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test/model work vs continuous-batching docs; no shared concrete bug." + "reason": "Mistral4 test fix vs CLIP-family layernorm typo mapping; different changes." }, { "left": "pull_request:44827", - "right": "pull_request:45170", + "right": "pull_request:45795", "accept": false, - "reason": "Mistral4 test/model work vs CLIP-like layernorm rename cleanup; different fixes." + "reason": "Mistral4 test/model changes vs docs for audio/video processors; unrelated." }, { "left": "pull_request:44827", - "right": "pull_request:45795", + "right": "pull_request:45655", "accept": false, - "reason": "Mistral4 fixes vs audio/video processor docs additions; unrelated." + "reason": "Mistral4 work vs config resolution order fix; different underlying issue." }, { "left": "pull_request:45170", "right": "pull_request:45899", "accept": false, - "reason": "CLIP-like layernorm rename fix vs continuous-batching docs; unrelated." + "reason": "Layernorm typo rename vs docs-only batching fast-path description; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 fixes vs model-creation skill PR; unrelated." + "reason": "Mistral4 tests/model code vs model-creation skill docs; not the same change." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "MobileLLM model PR vs model-creation skill PR; not the same artifact." + "reason": "New MobileLLM model vs model-creation skill docs; no shared code-path problem." }, { "left": "pull_request:45213", "right": "pull_request:45899", "accept": false, - "reason": "Skill/documentation PR vs continuous-batching docs update; unrelated topics." - }, - { - "left": "pull_request:44827", - "right": "pull_request:45655", - "accept": false, - "reason": "Mistral4 fixes vs config-resolution-order change in `modeling_utils`; different bugs." + "reason": "Skill docs for model creation vs continuous batching docs update; unrelated." }, { "left": "pull_request:45795", "right": "pull_request:45899", "accept": false, - "reason": "Docs for audio/video processors vs docs for decode fast path; separate topics." + "reason": "Docs for audio/video processors vs docs for decode fast path; different documentation topics." }, { "left": "pull_request:43823", "right": "pull_request:45795", "accept": false, - "reason": "MobileLLM model addition vs processor docs additions; unrelated." + "reason": "New MobileLLM model vs docs for processor components; unrelated." }, { - "left": "pull_request:45899", - "right": "pull_request:45946", + "left": "pull_request:43823", + "right": "pull_request:45899", "accept": false, - "reason": "Continuous-batching docs update vs RF-DETR release-org/docs update; unrelated." + "reason": "New MobileLLM model vs decode fast-path docs; no common underlying change." }, { "left": "pull_request:44827", "right": "pull_request:45922", "accept": false, - "reason": "Mistral4 test/model fixes vs vision-model LRU leak fixes; different code paths." + "reason": "Mistral4 test/model work vs vision-model cache-leak fix; unrelated." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs fix the same `fixed_cross_entropy` kwargs bug and both explicitly target issue 43240; they are direct variants of the same change." + "reason": "Same file, same function, same issue 43240, and both add the same `cross_entropy` kwargs support; these are duplicate fixes for the same code path." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The cluster-level summary is mostly grounded, but the PR ranking language is a bit too strong for the evidence. PRs 43251 and 43254 both explicitly target issue 43240 and both patch `fixed_cross_entropy`, so calling 43251 the \"most direct, minimal\" or \"clearest canonical duplicate\" overstates the distinction between them. A more conservative phrasing would treat 43251 as one reasonable representative of the same bugfix, not uniquely canonical." + "accept": true, + "feedback": "Grounded and conservative overall. The only accepted duplicate pair is well supported by the packet: both PRs modify `src/transformers/loss/loss_utils.py`, target issue 43240, and implement the same `fixed_cross_entropy` kwargs expansion. The other soft-edge pairs are rejected in a cautious, file- and change-based way without overclaiming a shared bug." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -17037,7 +16975,7 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "1fe460e518491a99854c211f8167d3193b1650b26771d2977068317dba890978", + "prepared_review_unit_hash": "cc55c2bca5fcae75723397ff187f7da679e99a1eb725387e982aee32259b7fcb", "component_budget": { "node_count": 2, "item_count": 2, @@ -17059,18 +16997,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs work in the same VibeVoice area, but they target different changes: #40546 is the initial VibeVoice implementation, while #42781 adds VibeVoice Realtime plus an acoustic tokenizer and generation updates. They share files because the second builds on the first, not because they fix the same concrete problem.", - "confidence": 0.93, + "summary": "These are related VibeVoice PRs, but they are not duplicates: PR 40546 is the base VibeVoice implementation, while PR 42781 adds a separate realtime variant with acoustic-tokenizer support and additional generation/feature-extraction work.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "#40546 is the more canonical PR because it appears to introduce the base VibeVoice model support and documentation, which the later realtime work extends.", + "canonical_pr_reason": "PR 40546 is the better canonical PR because it is the broader, foundational VibeVoice implementation and already covers the core model integration, docs, and Auto-class wiring.", "best_issue_reason": null, - "best_pr_reason": "#40546 is the best representative PR for the cluster since it is the foundational implementation rather than a follow-on variant with additional realtime/tokenizer changes.", + "best_pr_reason": "PR 40546 is the best representative because it establishes the main VibeVoice codepath; PR 42781 is a later draft focused on realtime-specific extensions rather than the same concrete change.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "Do not merge: the overlap is mostly shared VibeVoice infrastructure, but #40546 adds the core model integration while #42781 adds a realtime extension and acoustic tokenizer changes. They are related, but not the same underlying fix/change and would not plausibly be merged as one PR." + "reason": "They overlap on VibeVoice plumbing, but 42781 is a distinct realtime follow-up with extra acoustic-tokenizer and generation changes. It is not the same underlying fix/change and would not be safely merged as one PR." } ] }, @@ -17100,7 +17038,7 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "38bdc9e591b1e5e42418209f7e85e34d281cb381043beb880f88b0694867cbf4", + "prepared_review_unit_hash": "da5861b5337a1a214beebad06de6747884ec825e5e5e829d4975ef0e399c0a51", "component_budget": { "node_count": 2, "item_count": 2, @@ -17122,18 +17060,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests address unrelated features in different model families: TimesFM covariate forecasting vs DINOv3 image classification support. They should not be clustered as duplicates.", + "summary": "These two pull requests are unrelated feature additions: one adds covariate-based forecasting to TimesFM, the other adds DINOv3 image classification support. They do not appear to fix the same code-path problem or be mergeable into one PR.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No clear canonical PR: the PRs solve different problems in different code paths and are not mergeable into one duplicate representative.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "Neither PR is a good global representative for the cluster because they are unrelated; if forced, 41224 is the more self-contained model-support addition, but it does not subsume 40755.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different models and different functionality: TimesFM covariate regression/forecasting vs DINOv3 image classification. No shared concrete bug or change." + "reason": "Different models, different subsystems, and different goals: TimesFM covariate forecasting vs. DINOv3 image classification. No shared underlying bug/change." } ] }, @@ -17228,7 +17166,7 @@ "pull_request:41797|pull_request:45186", "pull_request:45186|pull_request:45630" ], - "prepared_review_unit_hash": "b0e5f961307823a970acd695077a2f82f99ef40bc163e199da19f7da5a8cd9ce", + "prepared_review_unit_hash": "6629ed434b5547738597a9e45f3ae5335e5c39796b1266be3a0ae38a01bff6f7", "component_budget": { "node_count": 3, "item_count": 3, @@ -17250,24 +17188,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three PRs are separate model-addition changes for different models, with overlapping boilerplate but no shared concrete bug or feature implementation.", + "summary": "These are three independent model-addition PRs (DeepSeek OCR, Isaac, and Kimi2-6). They share only the standard Transformers model-onboarding scaffolding, not the same underlying change.", "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: PRs 41797, 45186, and 45630 add different models (DeepSeekOCR, Isaac, and Kimi2-6), so they are not duplicates of the same change.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "If a representative PR is needed, 45630 is the strongest candidate because it has the broadest integration surface and highest review activity, but it is still a distinct model-addition PR rather than a duplicate.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:41797", "right": "pull_request:45186", "accept": false, - "reason": "Different model additions for different architectures; shared doc/auto-registration files are boilerplate and do not indicate the same underlying change." + "reason": "Different model integrations: DeepSeek OCR vs Isaac. Shared files are generic registry/docs/template updates typical of new model PRs, not the same code-path fix or mergeable duplicate change." }, { "left": "pull_request:45186", "right": "pull_request:45630", "accept": false, - "reason": "These add different models (Isaac vs Kimi2-6) with different codepaths and support files; overlap is only generic model-registration scaffolding." + "reason": "Different model integrations: Isaac vs Kimi2-6. The overlap is broad onboarding machinery and docs, but the concrete model code, configs, and tests are for unrelated models, so these are not duplicates." } ] }, @@ -17297,12 +17235,12 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "edf568e347e3a48c28d585d94851e49a1a4094c2dfdf333182d4e00652007efd", + "prepared_review_unit_hash": "b9ddd2a1be960ca3b58f11ff5f827b7d5486751a47a888b378a036a7264e13a4", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5933, + "serialized_chars": 5934, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -17310,7 +17248,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5933, + "serialized_chars": 5934, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -17319,18 +17257,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs both update GitHub Actions references, but they target different changes: one is a broad Node 24 compatibility sweep for workflow actions, the other upgrades docker/build-push-action in Docker image workflows. They overlap in area, not in the same concrete change.", - "confidence": 0.94, + "summary": "These are related only at the level of GitHub Actions maintenance, but they target different action families and different changes: PR 42975 upgrades core GitHub actions for Node 24 compatibility, while PR 42976 updates docker/build-push-action to v6 across Docker image workflows. They should not be merged as duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 42975 is the clearer canonical representative: it has the more specific stated goal (Node 24 compatibility) and a broader, more coherent set of workflow action upgrades.", "best_issue_reason": null, - "best_pr_reason": null, + "best_pr_reason": "PR 42975 is the best overall representative of the cluster because its scope is more self-contained and explicitly framed, whereas PR 42976 is a separate Docker action version bump.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Both touch GitHub Actions workflows, but they fix different things with different action families and scopes. They are not the same underlying bug/change and would not plausibly be merged as one PR." + "reason": "Different underlying changes: one upgrades core GitHub Actions for Node 24 compatibility, the other bumps docker/build-push-action to v6 in Docker image workflows. Same general maintenance area, but not the same fix/change." } ] }, @@ -17373,12 +17311,12 @@ "pull_request:44601|pull_request:45838", "pull_request:45490|pull_request:45838" ], - "prepared_review_unit_hash": "2827de19146996a595db26198bc46cc0eef5a96204b309ad0b1f41e526295317", + "prepared_review_unit_hash": "443824e7b861755b7db7690480fdf0858e946a73fac77320496b56b73d50fbbf", "component_budget": { "node_count": 7, "item_count": 7, "soft_pair_count": 9, - "serialized_chars": 23543, + "serialized_chars": 23541, "estimated_input_tokens": 5886, "estimated_eval_tokens": 12028 }, @@ -17386,7 +17324,7 @@ "node_count": 7, "item_count": 7, "soft_pair_count": 9, - "serialized_chars": 23543, + "serialized_chars": 23541, "estimated_input_tokens": 5886, "estimated_eval_tokens": 12028 }, @@ -17395,66 +17333,66 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are unrelated model-addition PRs, and one is a distributed pipeline-parallel loading change. The soft-similarity links are driven by shared scaffolding files (docs/auto registries) rather than the same bug or feature.", + "summary": "This cluster is a loose topical grouping of unrelated PRs: several independent model-addition PRs (ViT NEPA, Molmo, Molmo2, Lfm2 Audio, CTSM, PP-OCRv6 support) plus one distributed pipeline-parallel infrastructure PR. The apparent overlap is mostly boilerplate docs/auto-registry changes, not duplicate code changes.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: these PRs cover different models and one unrelated distributed-loading feature, so they are not duplicate variants of the same change.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 45838 is the strongest standalone PR in the set, with the broadest implementation and highest discussion activity, but it is not a duplicate of the others.", + "best_pr_reason": "PR 43451 is the strongest representative item here: it is a large, highly reviewed model-addition PR with substantial implementation/test coverage. It is not a duplicate of the others, but it is the most developed single PR in the cluster.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo. Shared auto/docs files are generic scaffolding, not the same fix." + "reason": "Different model families and code paths: ViT NEPA vs Molmo. Shared docs/auto-registration edits are standard boilerplate for model adds, not the same change." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo2. They do not target the same code-path problem." + "reason": "ViT NEPA and Molmo2 are separate model additions with different implementations, processors, and tests; no shared bug or identical feature change." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different model additions: ViT NEPA vs CTSM. Similar framework plumbing does not make them duplicates." + "reason": "ViT NEPA vs CTSM are unrelated model additions. Similar registry/docs edits do not indicate the same underlying change." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are related by name but are distinct model families with different files and APIs, so they are separate PRs." + "reason": "Molmo and Molmo2 are related by name but are distinct models with different files and functionality; this is not one fix/feature split across two PRs." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo and CTSM are unrelated model additions; shared registry/docs edits are not enough to merge them." + "reason": "Molmo and CTSM are different model stacks and different changes entirely." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 and CTSM are different model integrations with different implementations and tests." + "reason": "Molmo2 and CTSM are separate model additions with no shared concrete code-path problem." }, { "left": "pull_request:43973", "right": "pull_request:45838", "accept": false, - "reason": "Lfm2 audio model integration is unrelated to PP-OCRv6 text recognition; different feature areas and code paths." + "reason": "Lfm2 Audio adds an audio model, while PP-OCRv6 adds text-recognition models. Different modalities, code paths, and goals." }, { "left": "pull_request:44601", "right": "pull_request:45838", "accept": false, - "reason": "Pipeline-parallel loading support is a distributed-runtime change, not a model integration like PP-OCRv6." + "reason": "Pipeline-parallel loading support is core distributed infrastructure; PP-OCRv6 is a model-family addition. They are not the same bug or change." }, { "left": "pull_request:45490", "right": "pull_request:45838", "accept": false, - "reason": "CTSM and PP-OCRv6 are distinct model additions with separate architectures, files, and tests." + "reason": "CTSM and PP-OCRv6 are different model-family additions with distinct implementations and tests, not a duplicate pair." } ] }, @@ -17487,7 +17425,7 @@ "pull_request:43291|pull_request:44815", "pull_request:43665|pull_request:44815" ], - "prepared_review_unit_hash": "b70453f453129ab6e68b826ebb873264caaa11049f0a6c8dd33f5313bb127b2d", + "prepared_review_unit_hash": "98ae86de9abe81f55837c73daa31d2407544121d849be89f7be1c0b8f1f86cdd", "component_budget": { "node_count": 3, "item_count": 3, @@ -17509,30 +17447,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are only superficially similar via broad \u2018fix\u2019/test language. 43291 is a Whisper decode/tokenization bugfix, 43665 is test-only noise with failing asserts, and 44815 is an FP8/dequantization change in a different subsystem. None should be clustered as duplicates.", + "summary": "These three PRs are not duplicates. 43291 is a Whisper tokenizer/test fix, 43665 is mostly test-only noise, and 44815 is a substantive FP8/model-loading dequant change. The only apparent similarity is broad \u201cfix\u201d/test activity, which is not enough to merge them.", "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:43291 is the best canonical PR because it contains a coherent code fix in Whisper tokenization/decode behavior with matching test updates; the other PRs are either unrelated subsystem changes or not a real product fix.", + "canonical_pr_reason": "PR 44815 is the clearest representative of the cluster because it contains the only substantial code-path fix; the other two PRs are unrelated test/Whisper changes.", "best_issue_reason": null, - "best_pr_reason": "pull_request:43291 is the strongest representative of an actual fix. 43665 is not a substantive bugfix, and 44815 targets a different loading/FP8 path.", + "best_pr_reason": "PR 44815 is the strongest standalone PR: it fixes core model-loading / fine-grained FP8 behavior and includes targeted related changes, unlike the other two which are unrelated or test-only.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Different underlying changes: Whisper decode/tokenization bugfix vs. unrelated test-file edits with intentional failing asserts. They do not fix the same concrete issue and would not merge into one PR." + "reason": "Different areas and different kinds of changes: Whisper tokenizer logic vs CLIP/ViT test edits. No shared bug or mergeable combined fix." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Different code paths and subsystems: Whisper tokenizer decoding vs. core model loading / FP8 dequantization. The overlap is only the generic idea of \u2018fixing tests\u2019." + "reason": "Whisper timestamp decoding/tokenization is unrelated to FP8 dequant/model loading. Same generic \u201cfix\u201d label only." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "No shared bug or change: one is test-only placeholder edits, the other is a substantial FP8/dequantization fix. Not the same underlying problem." + "reason": "43665 is test-only noise for CLIP/ViT, while 44815 fixes FP8 dequant/model loading. Not the same underlying change." } ] }, @@ -17562,7 +17500,7 @@ "soft_pairs": [ "pull_request:43488|pull_request:43532" ], - "prepared_review_unit_hash": "0d71b2f08021595da24543f78fce041e5b85b4a89674e339d03fc10e698897be", + "prepared_review_unit_hash": "38c12097a507280dfb1541a78d110ddad94ad595ef053af4c08c70103f26b0c1", "component_budget": { "node_count": 2, "item_count": 2, @@ -17584,18 +17522,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two synthetic \"don't merge\" pull requests are present, but they target different changes: one is a repo-bot/version-format check, the other rewires CI/workflow/testing utilities. The shared Vit test file looks incidental, not a sign of the same underlying fix.", - "confidence": 0.91, + "summary": "These two open PRs are not duplicates: they target different changes, and the shared Vit test filename looks incidental. PR 43488 is a small setup/version/test-bot check, while PR 43532 changes CI workflows and testing utilities. Do not merge them together.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43532 is the more representative cluster member because it contains the broader, substantive code/workflow change set; PR 43488 is a narrow repo-bot/version-format check.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 43532 is the best representative PR for the cluster since it is the more complete and meaningful change, while 43488 is mostly a targeted formatting/check exercise.", + "best_pr_reason": "If one PR must be singled out as the cluster representative, 43532 is broader and has more context, but this is not a true duplicate cluster.", "soft_edge_verdicts": [ { "left": "pull_request:43488", "right": "pull_request:43532", "accept": false, - "reason": "Reject: these are different PRs with different goals and code paths. Shared Vit test filenames are incidental; one changes setup/version/deps, the other changes CI workflows and testing utilities." + "reason": "Different underlying changes: 43488 is about setup.py/dependency/version and a Vit test tweak, while 43532 modifies CI workflows, testing utilities, and related test files. Shared test file alone is not enough to treat them as the same fix." } ] }, @@ -17625,7 +17563,7 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "56ea3bb1ebc8d2213504cb35ef9f1bc70d2e61d4ba108da549a7e95f27aeb627", + "prepared_review_unit_hash": "625cd7798fc97138e880b6a7f06ec9f37d583258da591fcc9c2610aeec11d751", "component_budget": { "node_count": 2, "item_count": 2, @@ -17647,18 +17585,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate feature PRs for different segmentation pipelines: promptable concept segmentation vs promptable visual segmentation. They share docs and pipeline scaffolding, but they do not fix the same concrete code-path problem and should not be merged as duplicates.", - "confidence": 0.93, + "summary": "These are two separate pull requests adding different segmentation pipelines: one for Promptable Concept Segmentation and one for Promptable Visual Segmentation. They share some documentation and registration scaffolding, but the concrete pipeline behavior, model coverage, and test targets differ, so they should not be treated as duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 43613 is the more substantial standalone pipeline addition, with broader model/test coverage, but it is still a distinct feature from 43612 rather than a duplicate.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Different tasks and underlying functionality: concept segmentation vs visual segmentation. Shared infrastructure and similar filenames are not enough to treat them as the same change." + "reason": "Different underlying changes: one introduces Promptable Concept Segmentation, the other Promptable Visual Segmentation. Shared docs/auto-registration files are incidental and do not indicate the same code-path fix or a PR that could be merged as one." } ] }, @@ -17688,7 +17626,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "b7c118e220f53486a20cb3ff8a439ec8a08e48587d81d83fb3dbb7393f0406bd", + "prepared_review_unit_hash": "60e265cd84b773917a5e3e13cb74aa8d8f067bfbf56046e82da82442e4ba2ef5", "component_budget": { "node_count": 2, "item_count": 2, @@ -17710,18 +17648,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two unrelated open Trainer PRs share only the same file: one changes gradient-accumulation loss scaling in training_step, the other refactors signature-column filtering for custom/PEFT trainers. They target different code paths and are not duplicates.", - "confidence": 0.97, + "summary": "Two open PRs in the same trainer file, but they fix different trainer extension points: one changes gradient-accumulation loss scaling, the other extracts signature-column filtering for overrideability. They are related in subsystem, not duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43651 is the more concrete user-facing bug fix and the cleaner representative of this cluster, but it does not subsume the other PR.", + "canonical_pr_reason": "No canonical PR in this cluster: the PRs target different code paths and different customization hooks, so they should not be collapsed into one change.", "best_issue_reason": null, - "best_pr_reason": "PR #43651 is the strongest standalone change in the set because it fixes loss-scaling behavior directly in training_step; however, it is still unrelated to #43663.", + "best_pr_reason": "PR #43651 is the more representative concrete bug-fix style change in this cluster because it directly alters training-step loss normalization behavior; however, it still is not a duplicate of #43663.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Different functionality: #43651 changes gradient accumulation loss normalization, while #43663 adds an overridable helper for signature-column filtering. Same file only, not the same underlying fix." + "reason": "Different underlying problems: #43651 adds control over gradient-accumulation loss scaling, while #43663 refactors signature-column inspection for custom trainers/PEFT. Same file and similar theme, but not the same fix." } ] }, @@ -17753,12 +17691,12 @@ "pull_request:43656|pull_request:43842", "pull_request:43836|pull_request:43842" ], - "prepared_review_unit_hash": "161891150cdcd38989012f7a4669a3512ec5d85021040714e9e0e27ccab8d093", + "prepared_review_unit_hash": "1b3dc0ba6da7dd1bbec9b2b52eaa64f0839ce5ca8b062e4c7194cbbbf6b9f2d4", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 5992, + "serialized_chars": 5991, "estimated_input_tokens": 1498, "estimated_eval_tokens": 3252 }, @@ -17766,7 +17704,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 5992, + "serialized_chars": 5991, "estimated_input_tokens": 1498, "estimated_eval_tokens": 3252 }, @@ -17775,30 +17713,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All three PRs target the same serve.py import-time TypeAdapter/annotation NameError in the transformers CLI, with 43842 being the most focused fix. 43836 appears to bundle the same workaround into a broader serve.py cleanup/refactor, while 43656 is an earlier related annotation/future-import fix.", - "confidence": 0.79, + "summary": "These PRs cluster around a CLI `serve.py` annotation/runtime NameError issue involving `TypeAdapter` and related typing cleanup. The strongest duplicate pair is the two explicit `TypeAdapter`-focused fixes; the more generic annotation cleanup looks related but not clearly the same concrete change.", + "confidence": 0.67, "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:43842 is the clearest, most targeted fix for the reported TypeAdapter NameError when pydantic is absent, with minimal scope and explicit issue linkage.", + "canonical_pr_reason": "PR #43842 is the most precise and narrowly scoped fix for the reported `TypeAdapter` NameError when `pydantic` is absent, with a minimal change set in the affected code path.", "best_issue_reason": null, - "best_pr_reason": "pull_request:43842 is the best representative of the cluster because it directly addresses the concrete failure mode with the smallest, most precise change.", + "best_pr_reason": "PR #43842 is the best representative because it directly targets the specific runtime failure, uses the smallest focused patch, and is least likely to include unrelated typing cleanup.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", - "accept": true, - "reason": "Both are fixing the same import-time TypeAdapter/annotation NameError in src/transformers/cli/serve.py; 43842 is a narrower variant of the same underlying bug fix." + "accept": false, + "reason": "Both touch `src/transformers/cli/serve.py` and relate to annotation-name errors, but #43656 is broader future-annotations/type-hint cleanup while #43842 specifically fixes `TypeAdapter` availability; not clearly the same concrete bug fix." }, { "left": "pull_request:43836", "right": "pull_request:43842", "accept": true, - "reason": "Both target the same serve.py TypeAdapter NameError workaround and explicit shared issue; despite 43836 being broader, the core code-path problem appears the same." + "reason": "Both are clearly aimed at the same `TypeAdapter` NameError in `serve.py`, using stringized annotations/TYPE_CHECKING to avoid runtime evaluation; these look like overlapping fixes to the same underlying issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster read is grounded in the packet: all three PRs touch `src/transformers/cli/serve.py` and the titles/diffs support a shared import-time annotation/TypeAdapter NameError theme. The soft-edge accepts are also reasonably conservative because the changes are compatible and look like variants of the same fix rather than unrelated edits." + "feedback": "The clustering is grounded in the packet. The distinction between 43656 and 43842 is conservative, and the 43836/43842 pairing is reasonably supported by the shared `serve.py` TypeAdapter fix and shared issue target. The summary does not overclaim beyond the evidence." }, "evaluator_used": true, "retried": false, @@ -17924,8 +17862,8 @@ "pull_request:44066|pull_request:44086", "pull_request:44066|pull_request:44071", "pull_request:44018|pull_request:44068", - "pull_request:44018|pull_request:44066", "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", @@ -17987,367 +17925,367 @@ "pull_request:44044|pull_request:44129", "pull_request:44044|pull_request:44154" ], - "prepared_review_unit_hash": "e3a7e5a9bd5d78afc00931cf474a8df6ca3952a6a7e5ffbd6f8e847b6158adf8", + "prepared_review_unit_hash": "c7f05932f627e69748760f023cc4a63ce77ec28a38aee79e101cf9363479dc37", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84885, - "estimated_input_tokens": 21222, - "estimated_eval_tokens": 42700 + "serialized_chars": 84891, + "estimated_input_tokens": 21223, + "estimated_eval_tokens": 42702 }, "budget": { "node_count": 27, "item_count": 27, "soft_pair_count": 72, - "serialized_chars": 75080, - "estimated_input_tokens": 18770, - "estimated_eval_tokens": 37796 + "serialized_chars": 75088, + "estimated_input_tokens": 18772, + "estimated_eval_tokens": 37800 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a collection of model-specific output-tracing refactors. The only clear duplicate subgroups are GPTJ/CodeGen (44066, 44085, 44722) and GPT-Neo (44018, 44068); the rest share the same refactor theme but target different model code paths.", - "confidence": 0.81, + "summary": "This cluster is mostly a batch of model-specific refactors for standardized output tracing around issue #43979. The only clear duplicate-like overlaps are within the GPT-Neo pair and the GPT-J/CodeGen sequence; the rest are separate model refactors that share a theme but not the same concrete change.", + "confidence": 0.82, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44722 is the best representative for the GPTJ/CodeGen subgroup because it is a later overlapping revision of the same standardized output-tracing change and touches the same GPTJ/CodeGen files as the earlier variants.", + "canonical_pr_reason": "PR 44722 is the best canonical representative for the concrete GPT-J/CodeGen output-tracing refactor: it has the clearest implementation scope, explicit issue linkage, and subsumes the closely related GPT-J-only variants.", "best_issue_reason": null, - "best_pr_reason": "PR 44722 is the strongest single representative of the GPTJ/CodeGen duplicate subgroup because it is a later overlapping revision of the same fix, rather than a broader or clearly subsuming rewrite.", + "best_pr_reason": "PR 44722 is the strongest single PR in this cluster because it most clearly captures the underlying GPT-J/CodeGen output-tracing change and is the most complete version of that specific refactor.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different model families and code paths; only the output-tracing pattern is shared." + "reason": "Different model families and code paths; same issue target is not enough to make these duplicates." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Different model families and implementations; same refactor theme, not the same underlying change." + "reason": "CVT/FNet output tracing vs DeBERTa output tracing are separate model refactors." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Same GPTJ output-tracing change on the same file; 44066 just bundles an extra CodeGen edit, so they look like overlapping versions of one fix." + "reason": "Both touch the GPT-J output-tracing refactor on the same code path; 44066 just includes an added CodeGen companion change." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Both are standardized output-tracing refactors, but for different model backends and code paths." + "reason": "Different vision model refactors (ResNet/RT-DETR ResNet vs EfficientNet), not the same concrete change." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "Different model families and concrete changes; sharing the tracking theme is not enough." + "reason": "EfficientNet output tracing and GPT-J/CodeGen output tracing are unrelated concrete changes." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "Different model families and code paths; not the same concrete fix." + "reason": "GPT-J/CodeGen refactor vs EfficientNet refactor; only the shared theme is standardized tracing." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTaV2 are unrelated implementations of the tracing refactor." + "reason": "MobileNetV2 and DeBERTa are unrelated model implementations with different refactor targets." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "Different models and code paths; only a broad refactor pattern is shared." + "reason": "GPT-J/CodeGen output tracing and MGP-STR output tracing are different code paths." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "GPTJ/CodeGen versus MPT; not the same underlying bug/change." + "reason": "GPT-J/CodeGen refactor vs MPT refactor; same theme, different model behavior." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Both are GPT-Neo output-tracing refactors on the same file and look like overlapping versions of the same change." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44066", - "accept": false, - "reason": "GPT-Neo versus GPTJ/CodeGen; same architectural idea, but different concrete code paths." + "reason": "These are both GPT-Neo output-tracing refactors of the same model code path and look like the same underlying change." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPTJ/CodeGen versus GPT-Neo; same refactor theme, different model-specific fix." + "reason": "GPT-J/CodeGen and GPT-Neo are separate model refactors, so this is only thematic similarity." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet output tracing and GPTJ/CodeGen output tracing are unrelated concrete changes." + "reason": "ResNet output tracing and GPT-J/CodeGen output tracing are unrelated changes." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR versus GPTJ/CodeGen; different models and code paths." + "reason": "MGP-STR and GPT-J/CodeGen are different model-specific refactors." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "Different model families; the similarity is only at the refactor pattern level." + "reason": "ResNet and GPT-J are unrelated code paths despite the shared tracing theme." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are separate model implementations with different behavior changes." + "reason": "GPT-Neo and ResNet are different model implementations with distinct output handling." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "ResNet versus MPT; not the same underlying change." + "reason": "ResNet vs MPT is not the same underlying bug/change." }, { "left": "pull_request:44068", "right": "pull_request:44116", "accept": false, - "reason": "GPT-Neo versus Flaubert/XLM; similar tracing refactor, but different code paths." + "reason": "GPT-Neo output tracing and Flaubert output tracing are separate model changes." }, { "left": "pull_request:44068", "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo versus GPTJ/CodeGen; not the same model-specific fix." + "reason": "GPT-Neo and GPT-J/CodeGen are distinct model refactors, not one duplicate." }, { - "left": "pull_request:44059", - "right": "pull_request:44086", + "left": "pull_request:44071", + "right": "pull_request:44722", "accept": false, - "reason": "GPT2 versus MGP-STR; only the output-tracing pattern is shared." + "reason": "MPT and GPT-J/CodeGen are different concrete output-tracing changes." }, { - "left": "pull_request:44085", + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are separate models and separate changes." + }, + { + "left": "pull_request:44018", "right": "pull_request:44086", "accept": false, - "reason": "GPTJ versus MGP-STR; not the same concrete bug/change." + "reason": "GPT-Neo refactor vs MGP-STR refactor; same pattern, different code." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model refactors." }, { "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "ViLT versus Flaubert/XLM; unrelated model code paths." + "reason": "ViLT and Flaubert output-handling changes are unrelated concrete changes." }, { "left": "pull_request:44059", - "right": "pull_request:44722", + "right": "pull_request:44086", "accept": false, - "reason": "GPT2 and GPTJ/CodeGen are separate model families; not the same fix." + "reason": "GPT2 and MGP-STR output tracing are separate code paths." }, { - "left": "pull_request:44018", - "right": "pull_request:44071", + "left": "pull_request:44085", + "right": "pull_request:44086", "accept": false, - "reason": "GPT-Neo and MPT are different concrete implementations." + "reason": "GPT-J and MGP-STR are unrelated model implementations." }, { - "left": "pull_request:44017", - "right": "pull_request:44161", + "left": "pull_request:44059", + "right": "pull_request:44722", "accept": false, - "reason": "SegFormer and LongT5 are unrelated model code paths." + "reason": "GPT2 refactor and GPT-J/CodeGen refactor are not the same bug/change." }, { - "left": "pull_request:44017", - "right": "pull_request:44722", + "left": "pull_request:44018", + "right": "pull_request:44071", "accept": false, - "reason": "Same refactor theme, but different models and underlying changes." + "reason": "GPT-Neo vs MPT; only the broad tracing theme is shared." }, { "left": "pull_request:44017", - "right": "pull_request:44073", + "right": "pull_request:44161", "accept": false, - "reason": "SegFormer and VisualBert are different model families with separate fixes." + "reason": "SegFormer and LongT5 are different model-specific output-tracing refactors." }, { - "left": "pull_request:43997", - "right": "pull_request:44073", + "left": "pull_request:44017", + "right": "pull_request:44722", "accept": false, - "reason": "RegNet versus VisualBert; not the same change." + "reason": "SegFormer and GPT-J/CodeGen are unrelated concrete changes." }, { - "left": "pull_request:44073", - "right": "pull_request:44101", + "left": "pull_request:44068", + "right": "pull_request:44071", "accept": false, - "reason": "VisualBert versus Flaubert/XLM; different model-specific fixes." + "reason": "GPT-Neo and MPT are separate models with different implementation details." }, { - "left": "pull_request:44056", - "right": "pull_request:44161", + "left": "pull_request:44071", + "right": "pull_request:44076", "accept": false, - "reason": "MPNet and LongT5 are different code paths; similarity is only in the refactor pattern." + "reason": "MPT and ImageGPT are different model refactors." }, { "left": "pull_request:43997", - "right": "pull_request:44017", + "right": "pull_request:44073", "accept": false, - "reason": "RegNet and SegFormer are separate model implementations." + "reason": "RegNet and VisualBert are unrelated model-specific changes." }, { - "left": "pull_request:43997", - "right": "pull_request:44056", + "left": "pull_request:44017", + "right": "pull_request:44071", "accept": false, - "reason": "RegNet and MPNet are unrelated concrete changes." + "reason": "SegFormer vs MPT is only a shared architecture pattern, not the same change." }, { - "left": "pull_request:44071", - "right": "pull_request:44076", + "left": "pull_request:44073", + "right": "pull_request:44101", "accept": false, - "reason": "MPT and ImageGPT are different model families with different output-tracing code." + "reason": "VisualBert and Flaubert are different models and different output tracing updates." }, { - "left": "pull_request:44074", - "right": "pull_request:44101", + "left": "pull_request:44059", + "right": "pull_request:44071", "accept": false, - "reason": "TextNet and Flaubert/XLM are unrelated concrete changes." + "reason": "GPT2 and MPT changes are unrelated concrete refactors." }, { - "left": "pull_request:44059", - "right": "pull_request:44068", + "left": "pull_request:44056", + "right": "pull_request:44161", "accept": false, - "reason": "GPT2 versus GPT-Neo; same broad idea, but not the same bug/change." + "reason": "MPNet and LongT5 output tracing are separate code paths." }, { - "left": "pull_request:44059", - "right": "pull_request:44071", + "left": "pull_request:43997", + "right": "pull_request:44018", "accept": false, - "reason": "GPT2 and MPT are different model code paths." + "reason": "RegNet and GPT-Neo are unrelated implementations." }, { - "left": "pull_request:44068", - "right": "pull_request:44086", + "left": "pull_request:44017", + "right": "pull_request:44056", "accept": false, - "reason": "GPT-Neo versus MGP-STR; no shared concrete fix." + "reason": "SegFormer and MPNet are different model refactors." }, { - "left": "pull_request:44059", - "right": "pull_request:44018", + "left": "pull_request:44010", + "right": "pull_request:44071", "accept": false, - "reason": "GPT2 and GPT-Neo are separate implementations." + "reason": "SqueezeBert and MPT are different concrete changes." }, { - "left": "pull_request:44071", + "left": "pull_request:44073", "right": "pull_request:44161", "accept": false, - "reason": "MPT and LongT5 are unrelated model-specific refactors." + "reason": "VisualBert and LongT5 are separate model-specific output-tracing refactors." }, { "left": "pull_request:43997", - "right": "pull_request:44018", + "right": "pull_request:44056", "accept": false, - "reason": "RegNet and GPT-Neo are different model families." + "reason": "RegNet and MPNet are unrelated code paths." }, { "left": "pull_request:43997", "right": "pull_request:44161", "accept": false, - "reason": "RegNet and LongT5 are separate changes." + "reason": "RegNet and LongT5 are not the same underlying bug/change." }, { "left": "pull_request:44074", "right": "pull_request:44161", "accept": false, - "reason": "TextNet and LongT5 are unrelated concrete fixes." + "reason": "TextNet and LongT5 are different model refactors." }, { "left": "pull_request:44018", - "right": "pull_request:44129", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT2 are separate model-specific changes." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and SpeechT5 are different model code paths." + "reason": "MPNet and MPT are different implementations and not duplicates." }, { "left": "pull_request:44010", - "right": "pull_request:44073", + "right": "pull_request:44018", "accept": false, - "reason": "SqueezeBert and VisualBert are different implementations, not one duplicate fix." + "reason": "SqueezeBert and GPT-Neo are unrelated concrete changes." }, { "left": "pull_request:43997", - "right": "pull_request:44074", + "right": "pull_request:44017", "accept": false, - "reason": "RegNet and TextNet are unrelated model-specific changes." + "reason": "RegNet and SegFormer are different model refactors." }, { "left": "pull_request:44010", "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBert and DeBERTaV2 are different model families." + "reason": "SqueezeBert and DeBERTa are unrelated output-tracing changes." }, { "left": "pull_request:44044", "right": "pull_request:44074", "accept": false, - "reason": "DeBERTaV2 and TextNet are distinct code paths." - }, - { - "left": "pull_request:44066", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPTJ/CodeGen output-tracing change; 44722 is a later overlapping revision of that fix." - }, - { - "left": "pull_request:44085", - "right": "pull_request:44722", - "accept": true, - "reason": "Both are GPTJ output-tracing refactors on the same code path and look like overlapping versions of the same change." + "reason": "DeBERTa and TextNet are separate model-specific changes." }, { - "left": "pull_request:44085", - "right": "pull_request:44129", + "left": "pull_request:44010", + "right": "pull_request:44073", "accept": false, - "reason": "GPTJ and SpeechT5 are unrelated models and code paths." + "reason": "SqueezeBert and VisualBert are different code paths." }, { - "left": "pull_request:43996", - "right": "pull_request:44007", + "left": "pull_request:44044", + "right": "pull_request:44129", "accept": false, - "reason": "Different models and different concrete output-tracing implementations." + "reason": "DeBERTa and SpeechT5 are separate model refactors." }, { - "left": "pull_request:43996", - "right": "pull_request:44129", + "left": "pull_request:44044", + "right": "pull_request:44154", "accept": false, - "reason": "FNet/CVT versus SpeechT5; same theme, different fixes." + "reason": "DeBERTa and VITS are different concrete output-handling changes." }, { - "left": "pull_request:44044", - "right": "pull_request:44129", + "left": "pull_request:44056", + "right": "pull_request:44068", "accept": false, - "reason": "DeBERTaV2 and SpeechT5 are unrelated model-specific changes." + "reason": "MPNet and GPT-Neo are unrelated model refactors." }, { - "left": "pull_request:44044", - "right": "pull_request:44154", + "left": "pull_request:44018", + "right": "pull_request:44071", "accept": false, - "reason": "DeBERTaV2 and VITS are different model code paths." + "reason": "GPT-Neo and MPT are different concrete changes." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The overall cluster summary is mostly grounded, but the representative-selection reasons overstate the evidence. In particular, calling PR 44722 the \"latest, broadest revision\" and saying it \"subsumes\" earlier GPTJ-only variants is stronger than what the packet shows; 44066 already also touches GPTJ/CodeGen, so broadness/subsumption is not clearly established. Please weaken those claims to something like \"later overlapping revision\". The soft-edge accepts/rejects are otherwise conservative." + "accept": true, + "feedback": "The summary is grounded in the packet: most PRs are distinct model-specific output-tracing refactors, and the GPT-Neo and GPT-J/CodeGen overlaps are the only clear mergeable duplicates. The chosen canonical PR rationale is also supported by the diffs and issue linkage. Soft-edge verdicts are mostly conservative and do not overstate duplication." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -18382,70 +18320,70 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "02950f84b44dcb72788745ae241cc57238681725bb8f6e6eaeddd9b7fed5c5eb", + "prepared_review_unit_hash": "0869a19b82fbb11817c2090a714046ed5347e7b46aae7422b0a476f28958921a", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84885, - "estimated_input_tokens": 21222, - "estimated_eval_tokens": 42700 + "serialized_chars": 84891, + "estimated_input_tokens": 21223, + "estimated_eval_tokens": 42702 }, "budget": { "node_count": 8, "item_count": 8, "soft_pair_count": 6, - "serialized_chars": 17126, - "estimated_input_tokens": 4282, - "estimated_eval_tokens": 8820 + "serialized_chars": 17124, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a set of unrelated model-specific PRs that all apply the same framework-level output-tracing refactor pattern, but each targets a different model implementation. There is no issue artifact here, and none of the PRs are close enough to be duplicates of one another under a strict mergeability standard.", - "confidence": 0.88, + "summary": "These PRs are all instances of the same broad refactor pattern\u2014migrating individual model forwards to standardized output-capture utilities\u2014but they target different model implementations and code paths. As a duplicate cluster, they should stay separate; none of the soft edges are true duplicates under a strict PR triage standard.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the best canonical representative: it is clearly scoped, has an explicit issue target, and the diff shows a concrete output-tracing refactor in SpeechT5 rather than a vague title. It is the most fully specified example in the set.", + "canonical_pr_reason": "PR #44129 is the strongest cluster representative because it has the clearest problem statement, an explicit issue link, and a substantial output-tracing refactor in a model file consistent with the rest of the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the strongest exemplar for the cluster because it has the clearest linkage to a tracked issue and a substantive, model-specific implementation of the standardized output-capture refactor. It is more explicit than the other PRs while still fitting the shared theme.", + "best_pr_reason": "PR #44129 is the best single PR to represent the cluster since it most explicitly frames the standardized output-capture refactor and is easier to map back to the underlying initiative than the more model-specific variants.", "soft_edge_verdicts": [ { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "Both are output-tracing refactors, but they touch different models and different concrete code paths (FocalNet vs ImageGPT). They are similar in theme, not the same underlying change, so they should not be merged as duplicates." + "reason": "Same refactor theme, but different models and different code paths. These would not plausibly be merged into one PR without becoming an unrelated multi-model sweep." }, { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "VisualBert and TextNet are separate implementations with distinct forward/output logic. The common refactor pattern is too broad to treat as one duplicate PR." + "reason": "Both are output-tracing refactors, but they touch different architectures and files. Similar mechanics are not enough to make them duplicates." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "MPNet and VisualBert are different model code paths; this is a shared refactoring approach, not the same bug or change." + "reason": "MPNet and VisualBert are separate model implementations; the shared decorator/capture approach is a framework pattern, not the same underlying bug or change." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "These PRs both modernize outputs, but in different models with different internal hooks and forward signatures. They are not mergeable as one concrete fix." + "reason": "Distinct model-specific refactors with no shared concrete defect or identical code path." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model implementations. The overlap is only the standardized output-capture migration theme, which is too broad for duplicate triage." + "reason": "RegNet and SqueezeBert are unrelated model files. This is the same migration style, not the same change." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 modify different model paths and have different output-capture details. Same general refactor family, but not the same underlying PR change." + "reason": "These target different model internals and do not fix the same concrete issue; they only share the standardized output-capture motif." } ] }, @@ -18525,274 +18463,274 @@ "pull_request:44028|pull_request:44030", "pull_request:44029|pull_request:44030" ], - "prepared_review_unit_hash": "24ad967d8b82c6e201ed9c6fc6a9a47012286064178fff46812984f9e4d30c1e", + "prepared_review_unit_hash": "55567ac97fae8f84953354214a32da6800b280f4fe44ad1b21474baebfb34a32", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40175, - "estimated_input_tokens": 10044, - "estimated_eval_tokens": 20344 + "serialized_chars": 40172, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 }, "budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 40, - "serialized_chars": 40175, - "estimated_input_tokens": 10044, - "estimated_eval_tokens": 20344 + "serialized_chars": 40172, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All items are separate model-specific refactors to output tracing/can_return_tuple, not duplicates of one another. They share a pattern but target different architectures and code paths, so the cluster is only loosely related.", - "confidence": 0.9, + "summary": "These PRs share a common refactor theme, but they are separate model-specific changes rather than duplicates of one underlying bug or patch. I would not merge any soft edge pairs.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44003 is the most representative exemplar in the set: it is the largest change, touches two closely related Mamba-family files, and shows the common output-tracing refactor pattern more fully than the single-model PRs.", + "canonical_pr_reason": "PR 44003 is the most representative cluster anchor: it is a broader output-tracing refactor, spans two closely related model files, and introduces shared tracing helpers rather than a one-off local tweak.", "best_issue_reason": null, - "best_pr_reason": "PR 44003 is the best single representative for this cluster because it has the broadest scope and strongest evidence of the shared refactor style, while still being concrete and self-contained.", + "best_pr_reason": "PR 44003 is the best overall representative because it is the broadest and most infrastructure-oriented of the set, making it a better canonical example than the single-model refactors.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Both are output-tracing refactors, but for different models (timm_backbone vs upernet) and different forward paths." + "reason": "Different model families and files; same refactor theme only, not the same concrete bug/change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Different model families and code paths; only the refactor pattern is shared." + "reason": "Different models/code paths; only the output-tracing refactor pattern overlaps." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "timm_backbone and univnet are unrelated model implementations, so this is not the same underlying change." + "reason": "Unrelated model implementation; no shared underlying fix." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "These touch different architectures; 44003 is Mamba/FalconMamba, not the timm_backbone tracing path." + "reason": "Different architectures and code paths; not one mergeable fix." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Unrelated models and code paths; same refactor theme is not enough to merge." + "reason": "Different model families; same tracing refactor motif but no duplicate bug." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNetV1 and UnivNet are distinct model families with separate tracing changes." + "reason": "Separate models with different forward paths; not the same change." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "Different model code paths; both are tracing refactors but not the same bug/change." + "reason": "Different subsystems and model code paths; only a generic refactor similarity." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "MobileNetV1 vs DepthAnything are unrelated; the similar decorator/import changes are superficial." + "reason": "Different model families and filenames; not the same concrete fix." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "MobileNetV1 and VisionEncoderDecoder solve different forward-signature refactors in different subsystems." + "reason": "Encoder-decoder wrapper vs MobileNet model; no shared underlying bug." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "Different model families and output handling changes; not mergeable as one PR." + "reason": "Different model code paths; same refactor pattern only." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder and UnivNet are unrelated model paths despite similar refactor mechanics." + "reason": "Different models and output paths; not a duplicate change." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different architectures and forward implementations; no shared concrete bug." + "reason": "Different architectures and files; no same concrete bug/change." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "CLIP-like dual encoder vs Mamba-family stateful models are unrelated changes." + "reason": "Distinct model families; too broad a similarity to merge." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different model families; same output-tracing cleanup is not enough to treat as duplicate." + "reason": "Different model implementations and code paths; only same refactor style." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are unrelated model paths with separate changes to forwarding/output typing." + "reason": "Separate model code paths; not the same underlying issue." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Different concrete code paths in different architectures; not the same underlying fix." + "reason": "Mamba/FalconMamba vs UnivNet; unrelated concrete changes." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "UnivNet vs DepthAnything are unrelated refactors." + "reason": "Different models and outputs; no shared bug to coalesce." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "UperNet and RWKV are different subsystems; similar tracing pattern only." + "reason": "Different model families and forward implementations; not duplicates." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "UperNet segmentation code and Mamba sequence-model code paths are unrelated." + "reason": "Distinct code paths in different architectures; same meta-theme only." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Mamba/FalconMamba stateful tracing changes and SuperPoint feature extraction are different code paths." + "reason": "Different model families and files; not the same fix." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "Mamba-family refactor is unrelated to CodeGen attention/cache signature cleanup." + "reason": "Mamba/RWKV vs CodeGen are unrelated code paths; do not merge as duplicates." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "TimmBackbone and CodeGen are unrelated models; the overlap is only in output-tracing refactor style." + "reason": "Different model subsystems; only the tracing refactor label matches." }, { "left": "pull_request:44001", "right": "pull_request:44004", "accept": false, - "reason": "UnivNet and CodeGen change different forward signatures and return handling in unrelated models." + "reason": "Separate model implementations; not one concrete bug." }, { "left": "pull_request:44002", "right": "pull_request:44004", "accept": false, - "reason": "UperNet and CodeGen are unrelated architectures; not the same concrete bug/change." + "reason": "Different model families and forward logic; not duplicates." }, { "left": "pull_request:44000", "right": "pull_request:44004", "accept": false, - "reason": "Vision-text dual encoder and CodeGen are separate code paths; do not merge just because both refactor outputs." + "reason": "Vision/text dual encoder vs CodeGen are unrelated code paths." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "DepthAnything and VisionEncoderDecoder are unrelated model families with different tracing changes." + "reason": "Different model families; similar refactor pattern only." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "DepthAnything and SpeechEncoderDecoder are different code paths; similarity is only in pattern." + "reason": "DepthAnything vs SpeechEncoderDecoder are unrelated concrete changes." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "DepthAnything and SuperPoint are unrelated models, so this is not one shared fix." + "reason": "Different models and output structures; not the same fix." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "DepthAnything and RWKV address different forward/output handling in different architectures." + "reason": "Different architectures and files; no shared bug." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "DepthAnything and DPR are unrelated model implementations with separate refactors." + "reason": "DepthAnything vs DPR are distinct changes in different code paths." }, { "left": "pull_request:44026", "right": "pull_request:44029", "accept": false, - "reason": "VisionEncoderDecoder and RWKV are unrelated model families and code paths." + "reason": "Different encoder-decoder wrappers vs RWKV; not one underlying issue." }, { "left": "pull_request:44026", "right": "pull_request:44027", "accept": false, - "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are different encoder-decoder wrappers, but not the same underlying change." + "reason": "Different model wrappers and files; same tracing pattern only." }, { "left": "pull_request:44026", "right": "pull_request:44028", "accept": false, - "reason": "Different models and forward logic; same tracing refactor pattern only." + "reason": "Different models and concrete forward paths; not duplicates." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "VisionEncoderDecoder and DPR are unrelated subsystems with distinct output handling." + "reason": "Vision/audio encoder-decoder vs DPR are unrelated code changes." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "SpeechEncoderDecoder and SuperPoint are unrelated model code paths." + "reason": "Different models and implementations; no shared bug/change." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "SpeechEncoderDecoder and RWKV are different model families and cannot be merged as one fix." + "reason": "Speech encoder-decoder and RWKV are distinct code paths." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "SpeechEncoderDecoder and DPR are unrelated refactors." + "reason": "Different models and output mechanisms; not mergeable as duplicates." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are unrelated models; no shared concrete bug." + "reason": "SuperPoint vs RWKV are unrelated changes despite similar refactor wording." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "SuperPoint and DPR are separate implementations with different output-tracing changes." + "reason": "Different model families and code paths; not the same fix." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are unrelated code paths; the commonality is only the refactor style." + "reason": "RWKV and DPR are different implementations; only the output-tracing refactor theme overlaps." } ] }, @@ -18825,52 +18763,52 @@ "pull_request:44662|pull_request:45075", "pull_request:45075|pull_request:45401" ], - "prepared_review_unit_hash": "955779ca1b2881d208a9f03ebb43af7ac4e5d6a8899ec849a2d2eebf8535a607", + "prepared_review_unit_hash": "f924fe59d79e0dce6c7f557453056dd024ef64ccba3a6dc6bde8c7912eef62d9", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10912, - "estimated_input_tokens": 2728, - "estimated_eval_tokens": 5712 + "serialized_chars": 10913, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10912, - "estimated_input_tokens": 2728, - "estimated_eval_tokens": 5712 + "serialized_chars": 10913, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 }, "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, and Voxtral TTS). They share expected Transformers boilerplate, but they do not describe the same bug or change, so none should be merged as duplicates.", - "confidence": 0.97, + "summary": "These are three separate model-addition PRs (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They share some Transformers registry/docs plumbing, but they do not appear to fix the same underlying bug or implement the same concrete change, so none should be deduplicated together.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR as a duplicate target: the PRs implement different models and distinct code paths, with only framework-level files in common.", + "canonical_pr_reason": "PR #45075 is the best cluster representative: it is a substantial, non-draft model integration with conversion code, modeling, processing, tests, and docs. The other PRs are distinct model additions and not duplicates.", "best_issue_reason": null, - "best_pr_reason": "PR 45075 (DeepSeek-OCR-2) is the strongest standalone candidate: it is non-draft, has the highest activity, includes a converter plus model/image/processor code and tests, and appears the most complete of the three.", + "best_pr_reason": "PR #45075 is the strongest standalone PR in this set because it looks the most complete and self-contained, with explicit conversion and broad implementation coverage.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different model integrations (PenguinVL vs Voxtral TTS). Shared auto-registration and docs scaffolding are generic boilerplate, not the same underlying change." + "reason": "PenguinVL and Voxtral TTS are different models with different codepaths; overlap is limited to shared Transformers registry/docs files, which is not enough to treat them as the same change." }, { "left": "pull_request:44662", "right": "pull_request:45075", "accept": false, - "reason": "Different models and different implementations; overlap is only in Transformers registry/docs files and does not indicate one duplicate bug or mergeable PR." + "reason": "PenguinVL and DeepSeek-OCR-2 are unrelated model implementations. Shared auto-registration and docs files are expected for separate additions, not evidence of duplication." }, { "left": "pull_request:45075", "right": "pull_request:45401", "accept": false, - "reason": "DeepSeek-OCR-2 and Voxtral TTS are unrelated model additions with distinct converters, model classes, and test suites, so they are not duplicate fixes or changes." + "reason": "DeepSeek-OCR-2 and Voxtral TTS target different architectures and functionality. They only share common model-registration plumbing, so they should not be merged as duplicates." } ] }, @@ -18900,12 +18838,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "954038e39e1ace3a8515ad4186ac8db6363ec5e73711abea715f9a187750a0b5", + "prepared_review_unit_hash": "e30cbd0e07823ba3b76575d557f9fee68a9e9a003e77cfafdae5078174e677cf", "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 }, @@ -18913,7 +18851,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4407, + "serialized_chars": 4408, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -18922,18 +18860,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related kernel/module fusion efforts, but they target different layers and are not the same concrete change. The soft similarity is not enough to treat them as duplicates.", - "confidence": 0.92, + "summary": "The two PRs are thematically related around fusion, but they target different layers: one introduces a general module-fusion API, while the other wires up n-to-1 kernel fusion through KernelConfig and model-loading integration. They are not duplicates of the same concrete change.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45363 is the more concrete end-user feature: it wires n-to-1 kernel fusion through `KernelConfig`, model loading, and integrations. PR #44979 is a lower-level API foundation, but it is broader and less directly representative of the cluster\u2019s user-facing functionality.", + "canonical_pr_reason": "PR #45363 is the more concrete end-user feature: it connects KernelConfig to model loading and kernel-backed fusion across multiple integration points. It is the better representative for the cluster\u2019s practical fusion functionality.", "best_issue_reason": null, - "best_pr_reason": "PR #45363 is the best representative because it shows the full applied fusion workflow and touches the actual loading/integration path, whereas PR #44979 is mainly the underlying `Module Fusion API` scaffold.", + "best_pr_reason": "PR #45363 is the strongest overall representative because it has the clearest feature surface, explicit integration changes, and a user-visible workflow. PR #44979 is more of a foundational API piece than the full feature.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "They are related but not the same bug/change: #44979 introduces a generic module fusion API, while #45363 implements n-to-1 kernel fusion via `KernelConfig` and integration plumbing. They could coexist, not be merged as duplicates." + "reason": "Related area, but not the same underlying change. #44979 adds a standalone module fusion API in src/transformers/module_fusion.py; #45363 implements n-to-1 kernel fusion via KernelConfig and loader/integration plumbing. Different abstractions and code paths, so they should not be merged as duplicates." } ] }, @@ -18963,7 +18901,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "34359c38deba4ec3ba7c8f25380031908dce086b7132f72e9328f1a4b0b93e89", + "prepared_review_unit_hash": "4e3706aa6b9f0a8b7550d2da59a2e3e3f9abe18701f80fce9bbc8879c79d92e9", "component_budget": { "node_count": 2, "item_count": 2, @@ -18985,18 +18923,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests address the same Mistral regex AttributeError fix in the same code path, with 45317 appearing to be the more complete follow-up because it adds tests. They are close enough to be treated as duplicate/linked implementations of the same change.", - "confidence": 0.95, + "summary": "Both pull requests fix the same AttributeError in `_patch_mistral_regex` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; 45317 is the more complete version because it includes tests and clearer framing.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45317 is the better canonical PR because it targets the same bug fix as 45086 but is more complete: same code change plus test coverage and clearer problem statement.", + "canonical_pr_reason": "45317 is the better canonical PR because it targets the same bug as 45086 but adds test coverage and a clearer description of the fix.", "best_issue_reason": null, - "best_pr_reason": "PR 45317 is the best overall representative because it fixes the concrete AttributeError in _patch_mistral_regex and includes validation coverage, making it the most merge-ready version of the fix.", + "best_pr_reason": "45317 is the best representative PR for this cluster since it fully addresses the bug and includes validation, whereas 45086 is a smaller patch of the same change.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same underlying bug and same fix in the same file, both explicitly target issue 45081; 45317 is a fuller version of 45086 rather than a different change." + "reason": "Same underlying bug and same code path in `tokenization_utils_tokenizers.py`; the later PR is essentially a superset with added tests, so they could plausibly be merged into one fix." } ] }, @@ -19026,7 +18964,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "1b4d616a0b5d0fe1dd815be5471d9041f71708416233eb234b900d0ca874469a", + "prepared_review_unit_hash": "7f8ced45f61375ec7d764c00d673fdfe23dc51164b968d0bde7e31ce10247fc6", "component_budget": { "node_count": 2, "item_count": 2, @@ -19048,18 +18986,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs touch NemotronH internals, but they address different changes: one is a conversion/refactor to inherit GraniteMoeHybrid-compatible naming and auto/model wiring, while the other splits NemotronH into dense/sparse variants with docs and tests. They are related but not duplicates.", - "confidence": 0.91, + "summary": "Related NemotronH work, but not duplicates: PR #45115 is a conversion/refactor for legacy NemotronH weight mapping and GraniteMoeHybrid-compatible naming, while PR #45569 is a broader feature split that adds NemotronH Dense/Sparse variants, auto-mapping updates, docs, and tests.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45569 is the broader, more complete model-family change: it introduces the dense/sparse NemotronH variants, updates docs/auto mappings, and includes the model/test wiring needed for the split.", + "canonical_pr_reason": "PR #45569 is the broader umbrella change and the better representative of the NemotronH family update; it includes the new Dense/Sparse model structure and the surrounding integration work.", "best_issue_reason": null, - "best_pr_reason": "PR 45569 is the best representative of the cluster because it captures the main user-visible architectural change, whereas PR 45115 is a narrower conversion/mapping refactor.", + "best_pr_reason": "PR #45569 is the most complete and user-visible change in the cluster, so it is the best single PR to anchor on if one must be chosen.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "Not the same concrete fix. 45115 is about checkpoint conversion and refactoring NemotronH to GraniteMoeHybrid-style key/model inheritance; 45569 is about splitting NemotronH into dense/sparse model variants and updating docs/tests. Shared files reflect the same subsystem, but the code-path changes are different and would not naturally be merged as one PR." + "reason": "They touch some of the same NemotronH files, but the underlying changes are different: #45115 is a conversion/mapping refactor for inheritance from GraniteMoeHybrid, while #45569 introduces Dense/Sparse model variants plus docs/tests. They would not plausibly merge as one duplicate PR." } ] }, @@ -19089,7 +19027,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "695b58262f5594c6fdcdb4f224fb398bb46d47b1c152853919cd567f2deb05f1", + "prepared_review_unit_hash": "ec39b6b427c5404d80288dd59f606316fdede2caff9acbc8baac59d335402ae4", "component_budget": { "node_count": 2, "item_count": 2, @@ -19111,18 +19049,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related but not duplicates: one PR adds per-layer heterogeneous config support, while the other adds broader heterogeneous modeling support that builds on it. They share infrastructure, but they do not fix the same concrete change and would not be merged as one PR.", - "confidence": 0.91, + "summary": "Two related PRs in the heterogeneous-model workstream, but they are not the same change: #45333 adds per-layer heterogeneous config support, while #45332 builds on that with broader modeling/cache changes for heterogeneous models. They should stay separate.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45332 is the broader, more representative umbrella change: it introduces heterogeneous model behavior and also touches the config/cache/modeling plumbing needed to use per-layer settings end to end.", + "canonical_pr_reason": "PR #45332 is the broader umbrella change: it adds heterogeneous model support across cache/modeling code and includes the downstream consumers of the new per-layer config abstractions.", "best_issue_reason": null, - "best_pr_reason": "PR #45332 is the best representative of the cluster because it covers the larger concrete code-path change and includes the downstream modeling work, whereas #45333 is a narrower config-only subset.", + "best_pr_reason": "PR #45332 is the best representative of the cluster because it spans the full heterogeneous-model feature set, not just the config plumbing.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "Related infrastructure, but not the same underlying bug/change: #45333 adds per-layer config support, while #45332 adds heterogeneous model behavior and cache/model integration. They are complementary rather than duplicate PRs." + "reason": "Related work, but not the same concrete change. #45333 is config plumbing only; #45332 extends heterogeneous support into modeling/cache behavior and is broader than a mergeable duplicate." } ] }, @@ -19152,7 +19090,7 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "543745804729e37a6409cf51e5839987e53cedfc638964315689529b372104b4", + "prepared_review_unit_hash": "ef2d5f67b781a3ee50ac25c5911b63a0b1e43e5b3ae727afe5ed9244cde250ec", "component_budget": { "node_count": 2, "item_count": 2, @@ -19174,18 +19112,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both items are typing-related PRs, but they address different changes: one is a broad type-checking sweep across many transformer modules, while the other adds typing helpers/protocols centered on `_typing.py`, `modeling_utils`, and related integrations. They are related in theme, but not the same underlying change.", - "confidence": 0.9, + "summary": "Both PRs are typing-related, but they address different scopes: #45415 is a broad repository-wide type-checking/configuration pass, while #45425 adds specific typing protocols and helpers for model/PEFT-related code paths. They are related, but not the same change.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45415 is the broader, more representative typing cleanup across many files and the repo-wide type-checking config; it best captures the cluster theme.", + "canonical_pr_reason": "PR #45425 is the more concrete and representative typing change: it adds specific protocol/type definitions used by model-related code, rather than a broad type-checking sweep.", "best_issue_reason": null, - "best_pr_reason": "PR 45415 is the strongest cluster representative because it touches many transformer modules and establishes the broader type-checking pass, whereas PR 45425 is a narrower typing-infrastructure follow-up.", + "best_pr_reason": "PR #45425 is the best single representative artifact because it has a focused implementation goal and clear code-path impact, whereas #45415 is a wider type-checking cleanup across many files.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Both are about typing, but they are not the same concrete fix. 45415 is a broad type-checking sweep across many files; 45425 adds typing aliases/protocols and related adjustments in `_typing.py`/`modeling_utils`. They could not realistically be merged as one duplicate PR." + "reason": "Do not treat as duplicates: #45415 is a broad type-checking pass/config change across many modules, while #45425 introduces specific typing helpers for model/PEFT integration code. Related theme, different underlying change." } ] }, @@ -19224,7 +19162,7 @@ "pull_request:45858|pull_request:45926", "pull_request:45890|pull_request:45926" ], - "prepared_review_unit_hash": "d604d280d3d8335750c08249e8c192092a63549c6f2574fbce0aaf20b64894fb", + "prepared_review_unit_hash": "70f362b2fd0f75bfa7df5896e8e4a2b1e4ba3dc7143168794f4318e6bf2763c9", "component_budget": { "node_count": 6, "item_count": 6, @@ -19246,48 +19184,48 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs fall into two loose themes: model-parallel/device-placement fixes in model code, and XPU expectation updates in tests. Despite superficial similarity, they target different models and different code paths, so they should not be deduplicated together.", - "confidence": 0.83, + "summary": "These PRs are mostly separate model-specific fixes or test expectation updates. They share a broad theme of parallel/XPU-related work, but not the same underlying bug/change, so none of the soft edges should be merged.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45817 is the strongest representative of the cluster because it makes concrete runtime fixes in model code for Gemma4 model-parallel behavior, including no-split-module registration and device-aware masking logic.", + "canonical_pr_reason": "PR 45817 is the strongest representative: it contains a concrete code fix for Gemma4 model-parallel behavior, with an actual implementation change rather than only test-data updates.", "best_issue_reason": null, - "best_pr_reason": "PR 45817 is the best single PR to anchor the cluster because it addresses an actual model-parallel bug in implementation code, rather than only updating test expectations, and the fix is clearly tied to a concrete execution path.", + "best_pr_reason": "PR 45817 is the best single PR to represent the cluster because it is a substantive bug fix in model-parallel logic and is more central than the small test-only PRs.", "soft_edge_verdicts": [ { "left": "pull_request:45487", "right": "pull_request:45843", "accept": false, - "reason": "Both are model-parallel themed, but they fix different models (AltCLIP/ChineseCLIP vs Deimv2) with different placement logic; not the same underlying bug." + "reason": "Both are model-parallel fixes, but for different models and different failure modes: AltCLIP/ChineseCLIP device placement vs Deimv2 _no_split_modules propagation. Not the same underlying bug." }, { "left": "pull_request:45487", "right": "pull_request:45926", "accept": false, - "reason": "45487 is a runtime model-parallel fix, while 45926 only adds XPU test expectations for HyperCLOVAX; different artifact types and different models." + "reason": "AltCLIP/ChineseCLIP code fix versus HyperCLOVAX XPU expectation updates. Different models and one is test-only, so they should not be merged." }, { "left": "pull_request:45817", "right": "pull_request:45843", "accept": false, - "reason": "Both touch model-parallel handling, but Gemma4 vs Deimv2 have distinct code paths and fixes; too broad to treat as one bug." + "reason": "Gemma4 and Deimv2 have distinct model-parallel code paths and fixes. Same broad area, but not the same concrete bug." }, { "left": "pull_request:45858", "right": "pull_request:45890", "accept": false, - "reason": "Both are XPU expectation updates, but for different model test suites (HY-V3 vs Exaone4_5); they are not the same concrete change." + "reason": "Both only add XPU expectations, but they target different model test suites and different outputs. No shared underlying bug/change." }, { "left": "pull_request:45858", "right": "pull_request:45926", "accept": false, - "reason": "These are both test-only XPU expectation additions, but they target different models and different expected outputs, so they are not duplicates." + "reason": "Both are test expectation updates, but for different models and test cases. Too broad to treat as duplicates." }, { "left": "pull_request:45890", "right": "pull_request:45926", "accept": false, - "reason": "Same broad XPU-test theme, but different model families and different test files; no shared underlying bug or mergeable single PR." + "reason": "These are separate XPU expectation updates for different model suites, not the same concrete fix or change." } ] }, @@ -19317,7 +19255,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "c1f3df7b084c022ca74d3b7cc8a6a2de7d35ce15074cd1327030b4afaf14f9f3", + "prepared_review_unit_hash": "d2a708fe0009cc9081a1359d6893ba957a0123cdcc1ce169989d2d9b2f4304fa", "component_budget": { "node_count": 2, "item_count": 2, @@ -19339,18 +19277,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are about expert parallelism, but they fix different failure modes: #45548 addresses EP + DeepSpeed ZeRO-3 loading/serialization, while #45662 addresses EP + FSDP2 experts being overwritten by rank-0 broadcast. They share subsystem context, not the same underlying change.", - "confidence": 0.94, + "summary": "These two PRs both touch expert parallelism, but they address different failures: one is about preserving/loading EP-related config and enabling EP detection during DeepSpeed ZeRO-3 loading, while the other fixes FSDP2/rank-0 broadcast overwriting expert parameters during distributed loading. They are related subsystem-wise but not the same underlying bug or change.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: #45548 is a loading/config-serialization fix, while #45662 is a separate distributed sharding/broadcast fix. If forced to pick one representative, #45548 is the more self-contained loading-path change.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "#45548 is the better representative for the cluster because it is a narrower, concrete fix in the model-loading path; #45662 targets a different distributed execution bug and should not be merged with it.", + "best_pr_reason": "PR 45662 is the more concrete and self-contained bug fix: it targets a specific expert-parameter overwrite path during distributed loading and includes code changes directly addressing that failure mode.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "They are EP-related but not the same bug: one fixes accelerate/ZeRO-3 loading and config handling, the other fixes FSDP2 post-shard behavior and rank-0 broadcast overwriting experts. Different code paths and different failure modes." + "reason": "Different concrete bugs and code paths: 45548 focuses on config serialization/EP detection for DeepSpeed ZeRO-3 loading, while 45662 fixes expert tensors being overwritten during FSDP2/loading broadcast. They should not be merged as duplicates." } ] }, @@ -19380,7 +19318,7 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "2f72fb1efb0dd3cfbfd511f50ed41ebb8cebbca7855973d04681a5a1454c7c68", + "prepared_review_unit_hash": "ad207c371f962bbdddaf4437f9f0e89b21e205efadafb7dbf26f966a21593228", "component_budget": { "node_count": 2, "item_count": 2, @@ -19402,18 +19340,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two PRs address the same xdist/captured_info collision problem. PR 45645 is the broader, more complete fix and subsumes the narrower testing_utils-only change in 45639.", - "confidence": 0.82, + "summary": "Both PRs address the same xdist-related captured_info/debug-log collision problem in testing_utils; 45645 is the more complete variant because it also updates CI log collection to handle per-worker files.", + "confidence": 0.86, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the better canonical PR because it contains the core xdist-safe logging fix plus the CI workflow update needed to read multiple captured_info files, making it the more complete end-to-end solution.", + "canonical_pr_reason": "PR 45645 is the best canonical representative because it includes the core xdist-safe testing_utils fix plus the downstream CI workflow adjustment needed to consume multiple captured_info files.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the best PR for the cluster because it covers the shared bug fix in src/transformers/testing_utils.py and also updates the CI artifact collection path that depends on that change.", + "best_pr_reason": "PR 45645 is the strongest standalone candidate: it covers the same underlying bug as 45639 and adds the supporting CI handling, making it the more complete fix.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying captured_info xdist collision fix; 45645 appears to subsume 45639 and adds the complementary workflow handling for split artifact files." + "reason": "Accept: both PRs fix the same captured_info/xdist collision in the test debugging path, and 45645 looks like a broader follow-up of the same concrete change rather than a different bug." } ] }, @@ -19572,7 +19510,7 @@ "soft_pairs": [ "pull_request:45851|pull_request:45947" ], - "prepared_review_unit_hash": "019addd570fb95136768eab699dc505066bdeff479feaf177a310d3d4eb6bf0e", + "prepared_review_unit_hash": "2133b4f90d7396d57ff929fb3a603b0289caf70ccdb2be432cae7e415eb494d1", "component_budget": { "node_count": 2, "item_count": 2, @@ -19594,18 +19532,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both items are documentation PRs about chat templating, but they address different topics: one documents template storage/loading, the other adds prefill guidance for reasoning fields. They are related in subsystem only, not duplicates.", - "confidence": 0.93, + "summary": "Two open docs PRs in the chat templating area, but they address different topics: template file storage/loading vs. reasoning-field prefill behavior. They should not be deduplicated.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45851 is the stronger canonical representative because it is a more substantial chat-templating docs update and explicitly targets issue 45205.", + "canonical_pr_reason": "PR #45851 is the better anchor because it is explicitly linked to issue #45205 and has a focused, self-contained docs change about chat template storage/loading.", "best_issue_reason": null, - "best_pr_reason": "PR 45851 is the best single PR to keep as representative of the cluster because it has the explicit issue link and broader documentation scope.", + "best_pr_reason": "PR #45851 is the strongest representative: it has an explicit issue target, a clearer scoped documentation fix, and is more substantively connected to a concrete user-facing gap than #45947.", "soft_edge_verdicts": [ { "left": "pull_request:45851", "right": "pull_request:45947", "accept": false, - "reason": "They touch different docs and different user problems: template file storage/loading versus reasoning-field prefill behavior. Same area, but not the same underlying change." + "reason": "They are both chat-templating docs, but they describe different changes and different behaviors. #45851 explains on-disk chat template shapes; #45947 documents prefilling a reasoning field instead of content. Not the same underlying bug/change." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 1c2214ff88368b5d5ea72d129d6cd89a0627a4ad..016f10706614afc40420230e9e3a09eb4953241f 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260513t180036z", + "analysis_id": "hybrid-model-20260514t000044z", "archived_artifacts": { - "hybrid": "snapshots/20260513T180036Z/analysis-runs/hybrid-model-20260513t180036z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260513T180036Z/analysis-runs/hybrid-model-20260513t180036z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-05-13T18:06:50Z", + "published_at": "2026-05-14T00:09:20Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260513T180036Z", + "snapshot_id": "20260514T000044Z", "variant": "hybrid" } diff --git a/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.json b/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..a0c438ddecbcb44ffa4995454cb2441c7e6c30ce --- /dev/null +++ b/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.json @@ -0,0 +1,2076 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260514T000044Z", + "generated_at": "2026-05-14T00:09:18Z", + "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": 0.75, + "canonical_issue_number": 43824, + "canonical_pr_number": 43836, + "issue_numbers": [ + 43824 + ], + "pr_numbers": [ + 43836, + 43842 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "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 + ] + ] + } + ] + } + ] + }, + { + "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": [], + "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, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-43979-11", + "canonical_pr_number": 44007, + "duplicate_pr_numbers": [ + 43996, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "target_issue_number": 43979, + "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-44018-2", + "canonical_pr_number": 44068, + "duplicate_pr_numbers": [ + 44018 + ], + "target_issue_number": null, + "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-45081-3", + "canonical_pr_number": 45317, + "duplicate_pr_numbers": [ + 45086 + ], + "target_issue_number": 45081, + "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45561-3", + "canonical_pr_number": 45645, + "duplicate_pr_numbers": [ + 45639 + ], + "target_issue_number": 45561, + "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45647-3", + "canonical_pr_number": 45738, + "duplicate_pr_numbers": [ + 45737 + ], + "target_issue_number": 45647, + "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence." + } + ], + "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/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..a93285234a17039820f3ee4c97b484e0a6013294 --- /dev/null +++ b/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,19558 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260514T000044Z", + "generated_at": "2026-05-14T00:09:18Z", + "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-10000-400", + "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:43901|issue:44512", + "issue:41762|issue:43596", + "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": "49ac566543f39b872fa7fc45b0b8d1e5a252eca12eec40fe37b4bec356f92c44", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 62, + "serialized_chars": 46370, + "estimated_input_tokens": 11593, + "estimated_eval_tokens": 23442 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous cluster; most pairs are only superficially similar. The strongest duplicate groups are the ZeRO-3 Bert load failures, the Qwen3.5Moe rotary-embedding device/materialization reports, and the v5 AutoTokenizer token-ID regression reports.", + "confidence": 0.69, + "canonical_issue_reason": "issue:45920 is the broadest, most representative current report of the v5 AutoTokenizer token-ID regression across multiple model families; it subsumes the narrower Granite-specific report in issue:45812.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the best canonical issue because it describes the widest impact and the clearest reusable fix target among the accepted duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different models and different failures; not the same bug." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Unrelated hardware/backend errors with no shared code-path evidence." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different model families and different failure modes." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Both are dtype/device-style errors, but the model paths and bugs differ." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different model-specific failures; too broad to merge." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Unrelated model/runtime issues." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch vs ONNX export are unrelated." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Different subsystems and different concrete failures." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export and auto device mapping are unrelated." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs-related, but they target different removed v5 features." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export vs import error; not the same issue." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Different models and different runtime paths." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Same generic IndexError shape pattern, but not enough evidence of the same root cause." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "Same generic IndexError pattern, but distinct repros and model paths." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Different models and unrelated failure classes." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Different feature areas; no concrete shared bug." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Tokenizer detokenization bug vs LoRA merge/save mismatch are not the same defect." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Different model and different regression." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Both are shape-related, but they describe different training/runtime problems." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Different DeepSpeed errors and different affected code paths." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": true, + "reason": "Same Qwen3_5MoeTextRotaryEmbedding `inv_freq`/device-movement bug, surfaced via CPU offload vs `meta -> to_empty(cuda)` materialization." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "Both mention loading/runtime failures, but the model and failure mechanisms differ." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Different models and different numerical mismatch symptoms." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch vs trainer API signature mismatch are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation-checkpointing/FSDP issue vs import failure are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Different subsystems and different concrete errors." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs issues, but they concern different deprecated API text." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Both are ZeRO-3 loading errors, but the model contexts differ and the shared root cause is not established." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Different models and different runtime failures." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "Unrelated model/backend failure vs table QA crash." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure vs device mismatch are different bugs." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Cross-entropy/Liger-Kernel issue vs video StopIteration are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same v5 AutoTokenizer token-ID regression; issue:45920 is the broader multi-model report while issue:45812 is Granite-specific." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Checkpoint corruption vs load failure are different problems." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Different models and unrelated failure modes." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Same generic mask/index error wording, but not enough evidence of one underlying bug." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper model load/runtime issue vs table QA task crash are unrelated." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Same broad ZeRO-3 symptom family, but the model-specific repros are too different to merge confidently." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs cleanup, but they refer to different removed v5 surfaces." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Different model/runtime issues; no shared concrete bug." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "Different model families and unrelated behavior." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different model paths and unrelated backend errors." + }, + { + "left": "issue:44292", + "right": "issue:44918", + "accept": false, + "reason": "Different bugs with different model contexts." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger-Kernel cross-entropy issue vs SFT embedding unpacking are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same ZeRO-3 Bert loading failure with the identical `index 0 is out of bounds for dimension 0 with size 0` error." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The accepted duplicate pairs are well-supported by the titles, and the canonical issue choice for the v5 AutoTokenizer regression is reasonable from the packet." + }, + "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": 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-10000-400", + "nodes": [ + "issue:36010", + "issue:39186", + "issue:39290", + "issue:40990", + "issue:41093", + "issue:41762", + "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:45362|issue:45464", + "issue:43792|issue:45070", + "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:41093|issue:43596", + "issue:43541|issue:44589", + "issue:43054|issue:43994", + "issue:45542|issue:45800", + "issue:45759|issue:45902", + "issue:40990|issue:44960", + "issue:39186|issue:44918", + "issue:43716|issue:44918", + "issue:45632|issue:45698", + "issue:43404|issue:45005", + "issue:43645|issue:45399", + "issue:44530|issue:45005" + ], + "prepared_review_unit_hash": "58bcb3482a849cba68ac270f6c09d4875e4c8c24d3438ab773479384c83fa1af", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 37, + "serialized_chars": 37921, + "estimated_input_tokens": 9481, + "estimated_eval_tokens": 19218 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated Transformers issues are mixed together here; the soft links are generally just broad overlaps in model family, error type, or subsystem rather than the same concrete bug. I would not merge any of the soft pairs, and I\u2019d use the broad remote_code regression as the best representative issue.", + "confidence": 0.35, + "canonical_issue_reason": "issue:45020 is the broadest, most representative open bug report in the set: it describes a general regression where recent Transformers versions break models using `remote_code`, which is a common umbrella for several loading/configuration failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the strongest global representative because it is general, clearly scoped, and likely to cover a wider class of user-facing breakages than the more model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both mention sliding-window related config issues, but they affect different model families and symptoms; not the same concrete bug." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "One is a Torch version incompatibility, the other is a v5 Pydantic/config regression; too different to be duplicates." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "Different failure modes: DeepSpeed ZeRO-3 index error vs a Torch/Transformers compatibility issue." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Completely different bugs: detokenization regression vs table-question-answering crash." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both involve Qwen3.5 and chat paths, but the model size, API path, and failure mode differ; not clearly the same bug." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper runtime failure is unrelated to the Pydantic field regression in Transformers v5.4.0." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Different domains and code paths: Qwen NVFP4 loading vs table-question-answering crash." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Different models and symptoms; tied lm_head weights vs a corrupted checkpoint report." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma-related, but one is a missing config attribute and the other is a device mismatch at forward time." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Streaming chat failure on Qwen3.5 is unrelated to the MPS histogram/int error in a MoE model." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Different subsystems and symptoms: torch dynamo/grouped_mm tracing error vs MPS unsupported histogram op." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Both are dtype/runtime problems in MoE models, but the model families and exact code paths differ; too broad to merge." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Processor failure for Voxtral is unrelated to a template-node compilation error." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "Import path breakage for GenerationMixin is unrelated to the Gemma sliding_window_pattern issue." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are index/shape errors, but one is a mask-size mismatch and the other is a zero-size index during Gemma3 loading." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Version incompatibility and device placement mismatch are different bugs." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "Different model-support gaps: PaddleOCR-VL loading in vLLM vs GGUF gpt-oss architecture support." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing `pad_token_idx` in StableLmConfig is a model config regression, not the same as the PretrainedConfig/Pydantic issue." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Both are v5 regressions, but one is int4 memory/OOM behavior and the other is tied-weights handling on translation models." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Different MoE-related problems: torch dynamo grouped_mm failure vs tensor-parallel support issue for GPT-OSS." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Different model families and error paths; dtype mismatch under autocast is not the same as a streaming chat API crash." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "init_empty_weights constructor argument failure is unrelated to the PretrainedConfig/Pydantic regression." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF architecture support and Qwen3.5 streaming inference failure are distinct issues." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "A general remote_code regression is not the same as a Torch 2.4.1/Transformers 5.8.0 incompatibility." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Sliding-window config handling and Pydantic PretrainedConfig breakage are unrelated." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are indexing errors, but the triggering conditions and code paths differ substantially." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "Grouped_mm tracing failure and missing Float8 storage object are different low-level runtime issues." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": false, + "reason": "Both concern Siglip2 quality, but one is a text-embedding regression and the other is a generic nonsensical-output report; not enough to treat as the same bug." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard-without-TensorFlow backend detection is unrelated to Torch/Transformers version incompatibility." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Different bugs: config unwrapping for composite multimodal models vs uninitialized rotary embedding memory after meta/to_empty materialization." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "High perplexity on gpt-oss and a GLM5 issue are not the same underlying problem." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP weight-shape error and TRL SFT embedding unpacking failure are different training-path bugs." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Mistral image dtype mismatch is unrelated to Qwen3.5 embedding unpacking in TRL." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both involve remote/custom loading, but one is a cache-path collision and the other is loading the wrong custom module after save/load; not clearly the same bug." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but they concern different models and different manifestations of the problem." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash is unrelated to tied-weight handling on translation models." + } + ] + }, + "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": 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-10000-400", + "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:45399", + "issue:45507", + "issue:45520", + "issue:45663", + "issue:45701", + "issue:45800", + "issue:45859", + "issue:45902" + ], + "soft_pairs": [ + "issue:44360|issue:44512", + "issue:42915|issue:45399", + "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:43866|issue:44403", + "issue:41628|issue:45800", + "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:43782|issue:44960", + "issue:43881|issue:43883", + "issue:41762|issue:42915" + ], + "prepared_review_unit_hash": "988bfdee93c0550ba54fc2e49ba73a7cbb96493a04ea0b1f75ab3a9218ba7c10", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37480, + "estimated_input_tokens": 9370, + "estimated_eval_tokens": 18996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a grab-bag of unrelated Transformers issues; only one soft pair looks like a true duplicate: the non-persistent buffer corruption regression in v5. The rest share broad themes like loading, quantization, or model-specific failures but not the same concrete bug.", + "confidence": 0.79, + "canonical_issue_reason": "issue:43950 is the best canonical issue because it states the core regression clearly and broadly: `from_pretrained()` corrupting non-persistent buffers in Transformers 5.x. It is the most representative of the one clear duplicate pair with issue:44534.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43950 is the strongest standalone issue in this set: concrete, well-scoped, and it matches a near-identical duplicate entry.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "Different problems: a discussion about DSA indexer behavior vs docs referencing a removed run command." + }, + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Both involve model/runtime behavior, but one is a Qwen3Moe FP8 failure and the other is a flash-attn fallback gating issue." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to tied-weights handling in translation models." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "Model checkpoint loading failure is not the same underlying bug as tied-weights handling in v5." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "A cache-strategy typo is unrelated to the docs issue about a removed command." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Different code paths: Qwen3.5 embedding unpacking in TRL vs GraniteMoEHybrid calling an invalid method." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Both are model-specific loading/tokenization failures, but they are distinct models and distinct failure modes." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Near-identical regression: both report Transformers v5 filling non-persistent buffers with junk / corrupting `register_buffer(persistent=False)` state." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "Different failures: Qwen NVFP4 runtime error vs GraniteMoEHybrid invalid method call." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "SigLIP2 example/documentation errors are not the same as degraded text embeddings." + }, + { + "left": "issue:43643", + "right": "issue:43950", + "accept": false, + "reason": "Remote-code config field loss is a different bug than non-persistent buffer corruption." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Both mention dtype/position_ids-related problems, but they occur in different models and different failure mechanisms." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Trainer API keyword mismatch is unrelated to Gemma3 ZeRO-3 loading failure." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Checkpoint corruption and loading noise are not the same underlying issue." + }, + { + "left": "issue:41628", + "right": "issue:45800", + "accept": false, + "reason": "Import/export availability problems are unrelated to torch/Transformers version incompatibility." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Both concern buffer/initialization state, but one is a general regression and the other is a specific `inv_freq` uninitialized-memory bug." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "Bitsandbytes constructor incompatibility is unrelated to missing float8 storage support." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "Worse SigLIP2 text embeddings are not the same as version-dependent tokenization changes." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to missing `all_tied_weights_keys` on Molmo." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 and worse SigLIP2 text embeddings are related to the model family but are different bugs." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Both touch Qwen3.5/MoE, but one is Flash Attention misinterpreting 3D position_ids and the other is CPU offload compatibility." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_state KeyError is unrelated to Qwen3.5 rotary embedding uninitialized memory." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Bad tokenizer loading error and a model loading failure are not the same defect." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Gradient checkpointing with LoRA is unrelated to int4 quantization memory regression." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Both are SigLIP2-related, but one is text embedding quality and the other is HF-vs-JAX implementation discrepancy." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic load-time noise vs a specific failure to load one model are not the same bug." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "Remote-code field loss is not the same as broad recent version breakage for remote_code models." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Detokenization bug and import-utils flash_attn KeyError are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45237", + "accept": false, + "reason": "Detokenization on DeepSeek Coder is unrelated to GPT-OSS-20B AMD GPU support." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Both are loading-related, but one is a generic noise complaint and the other is a concrete model load failure." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "LoRA gradient checkpointing and Qwen3 MoE memory usage are different training-path issues." + }, + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Qwen3VL weight_only load error is unrelated to GLM5." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Different models and different failure causes: GLM-4V loading vs Molmo tied-weights attribute missing." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading failure and Qwen3Moe FP8 failure are distinct model-specific issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The selected canonical issue is well supported by the packet, and the one accepted soft pair (43950/44534) is a plausible duplicate based on closely matching titles. The remaining soft-edge rejections are conservative and do not overclaim equivalence." + }, + "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-10000-400", + "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:43827|issue:44509", + "issue:45081|issue:45399", + "issue:45020|issue:45356", + "issue:43404|issue:43950", + "issue:43742|issue:45399", + "issue:42915|issue:45020", + "issue:43257|issue:43866", + "issue:43638|issue:45823", + "issue:43792|issue:44220", + "issue:43872|issue:44291", + "issue:43940|issue:45313", + "issue:42915|issue:43872", + "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": "b42619d9cf91349ce6952ccd555bd4da59b9225c256f799424224dcc2af123a7", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 38066, + "estimated_input_tokens": 9517, + "estimated_eval_tokens": 19290 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issues; a few soft duplicates stand out around docs, initialization compatibility, DeepSpeed ZeRO-3 loading, mask-shape IndexErrors, and Qwen video regressions.", + "confidence": 0.67, + "canonical_issue_reason": "issue:45020 is the broadest, most representative current-regression report in the set, and it provides a reasonable anchor for comparing the more model-specific loading/tokenizer failures.", + "canonical_pr_reason": null, + "best_issue_reason": "Open, high-signal umbrella issue covering recent transformer-version regressions with real user impact; broader and more reusable as a triage anchor than the narrow per-model tickets.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "Different symptoms and code paths: MPS histogram/int handling vs Gemma4 PLE device placement." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "Both mention Qwen/LoRA, but one is a version-compatibility load failure and the other is a config warning; not the same bug." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "Importing a missing class and a bitsandbytes `_is_hf_initialized` constructor error are unrelated failures." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Remote-code breakage is a different problem from a Gemma4 device mismatch in forward." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "Tokenizer-mapping guard failure is not the same as the broader remote_code regression." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading failure for est-roberta is unrelated to flash-attn fallback gating." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Different model/runtime failures: autocast dtype mismatch vs device mismatch." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Same documentation regression: both still reference pipeline tasks removed in v5." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer regex patch crash is unrelated to flash-attn fallback selection." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Broad remote_code regressions and Kimi tokenizer codec handling are different root causes." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Weight tying in Mistral3 and non-persistent buffer corruption are distinct from each other." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "MobileLLM key error is unrelated to flash-attn fallback checks." + }, + { + "left": "issue:42915", + "right": "issue:45020", + "accept": false, + "reason": "Qwen3_moe LoRA compatibility between branches is a different issue from remote_code breakage." + }, + { + "left": "issue:43257", + "right": "issue:43866", + "accept": false, + "reason": "DeepSpeed weight conversion for Qwen3 MOE and a corrupted Ovis2 checkpoint are unrelated." + }, + { + "left": "issue:43638", + "right": "issue:45823", + "accept": false, + "reason": "Zero3/Bert empty-weights IndexError is not the same as Gemma4 device placement." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper not running is too broad to treat as the same bug as `_torch_extract_fbank_features()` without a clearer shared traceback." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Same `_is_hf_initialized` incompatibility triggered during model init/loading." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": true, + "reason": "Both report DeepSpeed ZeRO-3 failing to load Qwen3-family weights with missing parameters." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "LoRA/Qwen3_moe branch compatibility is unrelated to the bitsandbytes/init_empty_weights constructor error." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "A bad error message for Voxtral tokenization is not the same as an actual tokenizer load failure for a different model." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Different model families and different loading failures." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Unexpected trainer kwargs and mask/tensor shape mismatch are unrelated." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Both involve DeepSpeed, but one is weight conversion and the other is gradient averaging under GAS." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing ineffective in PEFT LoRA is not the same as a missing supports_gradient_checkpointing flag on a trust_remote_code model." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "NaN/SDPA fallback behavior in RLHF is unrelated to the flash_attention `s_aux=None` crash." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": true, + "reason": "Same mask-vs-index length mismatch IndexError, just reported with different concrete sizes." + }, + { + "left": "issue:43872", + "right": "issue:45538", + "accept": false, + "reason": "bitsandbytes init compatibility and CLIPTokenizer max length are unrelated." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Voxtral error-message quality and sleng-bert loading failure are different tokenizer issues." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": true, + "reason": "Both describe the same Qwen video-input regression in 5.3.0; 44560 is a specific manifestation." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "DeepSpeed training weight-shape error and void segmentation-map processing are unrelated." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Qwen2.5-VL rope scaling and Qwen3.5-MoE uninitialized-memory materialization are different bugs." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "GGUF gpt-oss support and Phi-tiny-MoE autocast dtype mismatch are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted soft-edge pairs are plausible duplicates or near-duplicates (docs v5 pipeline references, `_is_hf_initialized` init compatibility, ZeRO-3 weight loading failures, mask-shape IndexErrors, and the Qwen video regression), while the rejected pairs mostly distinguish different symptoms or root causes appropriately. No obvious overstatement of evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "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: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:45488", + "issue:45584", + "issue:45663", + "issue:45812", + "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:44451|issue:44488", + "issue:41720|issue:42915", + "issue:42491|issue:45276", + "issue:43708|issue:44075", + "issue:42915|issue:44756", + "issue:45488|issue:45812" + ], + "prepared_review_unit_hash": "6d5b54cadeeb026b9d33571c5677d245041865d7644ec0ebff17e4839df2c973", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37263, + "estimated_input_tokens": 9316, + "estimated_eval_tokens": 18888 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous Transformers regressions. I only found one clear duplicate pair (the v5 non-persistent-buffer junk regression) and one near-identical custom-model-init pair; the rest are too loosely related to merge.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 43644 is the clearest, most specific representative of the shared non-persistent-buffer regression and is better framed than its duplicate counterpart.", + "canonical_pr_reason": null, + "best_issue_reason": "As a local anchor for the one unambiguous duplicate subcluster, issue 43644 is the strongest fit; the broader cluster is too mixed for a single global canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but the bugs are different: FSDP2 KeyError in attention vs. device-mapping forward mismatch." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "Different code paths: batched AudioFlamingo token/embedding leakage vs Whisper empty-transcription behavior after align_special_tokens." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "Both mention Qwen/GGUF, but one is about gpt-oss architecture support and the other is a sliding_window bug." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Unrelated symptoms: non-persistent buffer corruption vs Kosmos2.5 OCR index error." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Different failures: Bamba torch.compile SDPA vs Qwen3.5 flash-attention position_ids misinterpretation." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Import error for Qwen2.5-VL is unrelated to the meta-device loading regression." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "These look like the same Transformers 5 custom-model initialization regression; one is the notebook-specific framing of the same failure." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "Different problems: GGUF gpt-oss support vs missing Float8 storage type during loading." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Qwen3Moe FP8 failure and meta-device loading are separate bugs." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Both are Qwen3.5/VL-related, but one is a video StopIteration issue and the other is TRL SFT embedding unpacking." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "3d_parallel.py runtime failure is not the same bug as GGUF gpt-oss support." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "Importing GenerationMixin and Qwen3Moe FP8 are unrelated regressions." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping add-new-model-like failure is unrelated to Gemma4 device mismatch." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": false, + "reason": "Same model name, but the titles only show broad bad-output symptoms; not enough to confirm the same concrete bug path." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized flag breakage is unrelated to Kosmos2.5 OCR indexing." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different failure modes: video StopIteration vs mask/tensor shape IndexError." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Strict config loading and glm4v config validation are separate issues." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "Zero-3/BertModel index error is not the same as the tokenizer-mapping add-new-model-like bug." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Both involve loading models, but the specific failures differ too much to treat as duplicates." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer mapping failure and mask-shape IndexError are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "gpu+cpu offload dequantized-model loading is a different issue from disabling mmap on Strix Halo." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Near-verbatim duplicate: both report the same Transformers 5 non-persistent-buffer junk regression." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Both mention SigLIP2, but one is output corruption and the other is documentation example errors/quantization failure." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "GLM5 and Qwen3.5 flash-attention are different model-specific bugs." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "LoRA qwen3_moe hf4.x vs BitNet packed-weight unpacking are unrelated regressions." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "DeepSpeed ZeRO-3 language_model weight loading and incorrect visual encoder keys are different Qwen3.5 problems." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are loading failures, but the titles do not establish the same concrete root cause." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "A800 cudaErrorAssert on auto device mapping is unrelated to Qwen3Moe FP8 failure." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "LoRA checkpoint compatibility and gemma4 resize_token_embeddings behavior are different bugs." + }, + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Trainer resume/max_steps accounting and SGD args not being used are separate training issues." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3Moe FP8 and Strix Halo mmap OOM are unrelated." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Tokenizer pre-tokenizer override regression and Granite AutoTokenizer wrong IDs are different tokenizer issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall: the cluster is indeed mostly heterogeneous, and the one clear duplicate pair (43644/44534) is strongly supported by the titles. The custom-model-init pair (43645/43646) is also a reasonable conservative accept from the available evidence. No PR-pair mergeability issue applies here." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43335", + "issue:43381", + "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:44928", + "issue:44960", + "issue:44991", + "issue:45216", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45362", + "issue:45406", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45704", + "issue:45759", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:43854|issue:43931", + "issue:39692|issue:43994", + "issue:43257|issue:43854", + "issue:44488|issue:44991", + "issue:42491|issue:43931", + "issue:43644|issue:43645", + "issue:43856|issue:43873", + "issue:45362|issue:45406", + "issue:43856|issue:44387", + "issue:44910|issue:45588", + "issue:42491|issue:45292", + "issue:36010|issue:39404", + "issue:45276|issue:45335", + "issue:44164|issue:44291", + "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:43854|issue:45216", + "issue:44292|issue:44912", + "issue:44387|issue:44756", + "issue:44743|issue:44960", + "issue:43957|issue:44756", + "issue:45684|issue:45698", + "issue:45276|issue:45292", + "issue:43381|issue:44928", + "issue:44164|issue:44756", + "issue:42947|issue:43381" + ], + "prepared_review_unit_hash": "3b3dc9fc5d6d43c85d50d4fcbc36da64c3e74b472910bfb9fca69a83f083567b", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 36, + "serialized_chars": 38614, + "estimated_input_tokens": 9654, + "estimated_eval_tokens": 19564 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose topical grouping of unrelated Transformers issues. The only clear duplicate-like group is the resize_token_embeddings regression affecting Gemma4/T5Gemma embedding modules; the rest should stay separate.", + "confidence": 0.77, + "canonical_issue_reason": "issue:45276 is the best canonical issue for the duplicate set because it describes the broader resize_token_embeddings failure and already covers multiple affected embedding paths (embed_tokens_per_layer and output_embeddings).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45276 is the strongest representative issue in this cluster: concrete repro, broader scope than the other resize_token_embeddings reports, and it best captures the shared underlying bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Both are model-loading failures, but they affect different models and failure modes; no evidence of the same underlying bug or a single fixable code path." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Both mention SigLIP2, but one is a documentation/example problem while the other is an inference correctness issue; not the same bug." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different models and different symptoms (Qwen3 MoE conversion/loading vs GLM-4.7 unit-test loading failure)." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but for different models and likely different tokenizer/config breakages; too broad to merge." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Distinct Qwen3 MoE training/load issues with different triggers and error classes." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Both are Transformers 5.0.0 regressions, but one is about non-persistent buffers and the other about custom model init in Jupyter; not the same bug." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Both involve memory/quantization/offloading concerns, but the concrete failure modes and code paths differ." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different models and different failures (Qwen3.5 chat crash vs Gemma4Processor missing attribute)." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both are memory-related, but one is training efficiency and the other is int4 quantization reserved-memory OOM; not the same bug." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Different attention/backend crashes with different causes and affected paths." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Both mention related Qwen/Gemma-style regressions, but they are not the same underlying bug or code path." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to Whisper pipeline return_language behavior." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": true, + "reason": "Both report the same resize_token_embeddings regression: embeddings are resized but not propagated to model-specific input/output embedding modules." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Both concern loading/saving, but one is extra_state handling and the other is init_empty_weights argument compatibility; different fixes." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both involve tied lm_head weights, but they are different model families and different manifestations." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Different Qwen3 device-map crash vs AudioFlamingo batch leakage; not the same underlying issue." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Both are vocab/tokenizer-related, but one is token IDs outside vocab size and the other is model loading failure; not a duplicate." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "SigLIP2 documentation/example errors are unrelated to GLM5 loading." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Both involve compatibility regressions, but one is Torch version incompatibility and the other is meta-device loading breakage." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing with PEFT LoRA is unrelated to a T5 apex RMSNorm memory leak." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states KeyError is unrelated to AutoModelForCausalLM.from_config text_config unwrapping." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Different model bugs: SwitchTransformers sparse-layer construction vs Gemma4 attention/FSDP state handling." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Related to memory use, but offloading/quantization behavior is not the same as int4 reserved-memory regression." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Gemma4 FSDP2 state rebuild bug is unrelated to Qwen3_5MoeTextRotaryEmbedding CPU offload incompatibility." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "init_empty_weights argument handling and disabling mmap on Strix Halo are different loading problems." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo batch leakage and Gemma4 PLE device mismatch are unrelated model-specific bugs." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Different models and failure modes: unit-test load failure vs incorrect saved checkpoint contents." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Both are quantization loading issues, but for different models/quantization formats and likely different code paths." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Both touch memory behavior, but one is quantization OOM under Transformers 5.x and the other is mmap on Strix Halo; not the same bug." + }, + { + "left": "issue:44743", + "right": "issue:44960", + "accept": false, + "reason": "Qwen modular recurrent-state reset bug is unrelated to GLM5 model-loading." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading regressions and mmap-on-Strix-Halo OOM are different issues." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save_pretrained/from_pretrained custom-model workflows, but one is file permission propagation and the other is loading the wrong custom module." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same resize_token_embeddings regression; both report that resizing does not update output embeddings." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Gradient checkpointing in eval mode is unrelated to the Qwen3.5 3D position_ids / SDPA fallback issue." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to mmap/OOM behavior." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Gradient checkpointing with PEFT LoRA is a different bug from checkpointing being disallowed in eval mode." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only duplicate-like cluster with strong title-level overlap is the resize_token_embeddings group (45276/45292/45335), and the canonical issue choice is plausible from the broader title scope. The other soft-edge rejections are appropriately cautious and do not overstate shared root causes." + }, + "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-10000-400", + "nodes": [ + "issue:32090", + "issue:36032", + "issue:39404", + "issue:43257", + "issue:43299", + "issue:43425", + "issue:43452", + "issue:43596", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43856", + "issue:43901", + "issue:43906", + "issue:43931", + "issue: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:44991", + "issue:45081", + "issue:45084", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45701", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "issue:44479|issue:44877", + "issue:43646|issue:43708", + "issue:45588|issue:45663", + "issue:45292|issue:45335", + "issue:43645|issue:44164", + "issue:43257|issue:44863", + "issue:44402|issue:44451", + "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:43901|issue:44509", + "issue:39404|issue:44912", + "issue:44265|issue:44898", + "issue:44479|issue:44805", + "issue:43994|issue:44079", + "issue:43299|issue:43931", + "issue:39404|issue:44671", + "issue:43425|issue:44918", + "issue:44589|issue:45084", + "issue:44164|issue:45759", + "issue:36032|issue:43452", + "issue:43596|issue:44683", + "issue:44671|issue:44991", + "issue:32090|issue:39404", + "issue:43856|issue:44756", + "issue:44991|issue:45081" + ], + "prepared_review_unit_hash": "2e4cef05bb96ac5f430ade1e096c3acbf1de25e463daa8cdc92db5d8b1056216", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 38057, + "estimated_input_tokens": 9515, + "estimated_eval_tokens": 19286 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous Transformers v5 regression reports, with one clear duplicate pair around `resize_token_embeddings` behavior. Most other pairs only share a broad area (model loading, tokenizers, multimodal configs, docs) and should stay separate.", + "confidence": 0.43, + "canonical_issue_reason": "`issue:43611` is the broadest representative of the main theme in this cluster: Transformers v5 breaking model loading / initialization paths. It is not a perfect fit for every item, but it best captures the dominant regression pattern here.", + "canonical_pr_reason": null, + "best_issue_reason": "`issue:43611` is the best single issue to anchor the cluster because it describes a general v5 loading regression rather than a narrow model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Both are v5-era multimodal loading regressions, but they hit different models and different failure modes; not the same bug." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization in notebooks is unrelated to `resume_from_checkpoint` / `max_steps` calculation." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Different code paths and failures: `s_aux=None` crash in flash attention vs Gemma4 FSDP2 `shared_kv_states` rebuild issue." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same underlying `resize_token_embeddings` propagation bug: output/decoder embeddings are not updated consistently." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Both touch model lifecycle code, but notebook custom-model init and `extra_state` handling are distinct problems." + }, + { + "left": "issue:43257", + "right": "issue:44863", + "accept": false, + "reason": "Different models and different loading failures; not a shared concrete bug." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "One is a tokenizer vocab-size mismatch, the other is a general model load failure for a different checkpoint." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Autocast dtype mismatch and meta-device loading breakage are different failure modes." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version incompatibility is not the same bug as the Phi MoE autocast dtype mismatch." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Float8 storage lookup failure and mmap OOM on Strix Halo are unrelated." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex attention failure and Gemma4 device mismatch are different code paths." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline `return_language` regression is unrelated to the broad tokenization-change report." + }, + { + "left": "issue:43931", + "right": "issue:44863", + "accept": false, + "reason": "Both are model-loading issues, but they involve different model families and different root causes." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "An isolated reproduction of a prior issue is not the same bug as the Qwen3.5 paged-attention crash." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Torch 2.10 incompatibility and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both involve v5 breaking custom-model flows, but one is generic base-model loading and the other is notebook-specific initialization." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch compatibility and Mistral-3 dtype mismatch are different issues." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttention cache crash and template compilation error are different failures." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs problems, but they cover different deprecated/removed pipeline task mentions." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline behavior and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export / `torch_compilable_check` is a different code path from Perceiver interpolation failure." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video input regression and mask-shape IndexError do not share the same underlying bug." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Siglip2 bad outputs and `ModelOutput` key handling are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both mention Qwen3-VL loading, but one is MoE conversion and the other is checkpoint shape mismatch; not enough evidence of one concrete bug." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper return_language regression and CamemBERT masked LM behavior are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch incompatibility and Qwen3.5 input-embedding unpacking are different problems." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Float8 storage lookup and template compilation failure are unrelated." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "`extra_state` save/load handling and `from_config` text_config unwrapping are different issues." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5 tokenizer special-token conflict and GGUF file handling are unrelated tokenizer/model-loading bugs." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "DeepSpeed ZeRO3 init IndexError and compiled flex attention failure are different code paths." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "CamemBERT masked-LM predictions and tokenizer loading for est-roberta are not the same bug." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast `NoneType` error and Whisper pipeline regression are unrelated." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3 MoE memory usage and Strix Halo mmap OOM are different performance/resource issues." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer load failure and Mistral regex patch crash are related only broadly by tokenizer code, not the same concrete defect." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet and stays appropriately conservative. The broad characterization as mostly heterogeneous v5 regression reports is supported, and the single accepted duplicate pair (45292/45335) looks plausibly like the same resize_token_embeddings bug. The other soft-edge rejections do not overclaim sameness." + }, + "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-10000-400", + "nodes": [ + "issue:34567", + "issue:36032", + "issue:40444", + "issue:40990", + "issue:42915", + "issue:43262", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43421", + "issue:43452", + "issue:43482", + "issue:43540", + "issue:43541", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43883", + "issue:43994", + "issue:44164", + "issue:44206", + "issue:44220", + "issue:44451", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44530", + "issue:44534", + "issue:44610", + "issue:44661", + "issue:44683", + "issue:44737", + "issue:44811", + "issue:44849", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45507", + "issue:45584", + "issue:45859" + ], + "soft_pairs": [ + "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:34567|issue:40444", + "issue:44811|issue:45584", + "issue:36032|issue:42915", + "issue:43644|issue:43646", + "issue:44451|issue:44987", + "issue:43452|issue:43482", + "issue:43541|issue:45084", + "issue:42915|issue:44530", + "issue:44849|issue:45325", + "issue:43317|issue:43482", + "issue:36032|issue:43645", + "issue:44987|issue:44991", + "issue:43883|issue:44991", + "issue:43404|issue:44466", + "issue:45084|issue:45507", + "issue:43262|issue:43377", + "issue:43994|issue:44610", + "issue:43646|issue:44534", + "issue:43322|issue:43540", + "issue:43645|issue:44534" + ], + "prepared_review_unit_hash": "a48c46421f835c57885b7444fe17437c40aa4cc5313d830b46738e394145001f", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37328, + "estimated_input_tokens": 9332, + "estimated_eval_tokens": 18920 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mix of unrelated issue reports, with only a couple of clear duplicate pockets around Transformers v5 regressions: non-persistent buffer serialization and custom model initialization. I would keep the non-persistent-buffer bug as the main canonical issue.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 43644 is the most concrete representative of the duplicated v5 buffer-regression cluster, and its wording is essentially repeated by 44534.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43644 is the best single issue to anchor the cluster because it is specific, directly duplicated, and captures the most obvious repeated failure mode.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Different models and symptoms: XLNet CPU device-placement bug vs Qwen3.5 MoE CPU-offload incompatibility." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "Unrelated areas: save/from_pretrained extra_state handling vs video-input regression." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon allocator TypeError and audio chat-template sampling-rate bug are unrelated." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-loading complaints, but for different models and different failures; not the same concrete bug." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer mapping helper failure and compiled flex_attention regression are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity regression and runtime special-token/post-processor update bug do not share the same code-path." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both mention loading failures, but they involve different model families and different loader paths." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer serialization." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "Same broad audio subsystem, but the titles point to different feature-extraction failures." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Mistral-3 image dtype mismatch and template-compilation error are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Different model-loading breakages with no clear shared bug." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "trust_remote_code config field loss and Qwen3.5 cache/group-type crash are unrelated." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "Trainer token accounting and multimodal iterable-dataset failure are unrelated." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both are Whisper-related, but one is batch_decode skip_special_tokens and the other is empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "Tokenizer method-name conflict and Qwen3Moe FP8 failure are unrelated." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Non-persistent buffer junk and custom model initialization are adjacent v5 regressions, but not the same concrete bug." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Both are loading regressions, but for different models and different failure modes." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both involve GGUF loading, but one is a generic from_pretrained breakage and the other is a specific Qwen2.5-GGUF v5 failure; not enough to merge strictly." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo grouped_mm tracing failure and template compilation error are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Different Qwen3-family bugs: FP8 config failure vs linear_attention cache crash." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Hidden-states bug in Qwen3.5 and Qwen2.5-VL rope-index scaling bug are different code paths." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "Dequantized-model device_map offload failure and Qwen2.5-GGUF loading regression are not the same bug." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer special-method conflict and custom model initialization regression are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are version-gated loading failures, but they affect different artifacts and loader paths." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Different model-loading issues with no shared concrete root cause." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Mistral3 lm_head tying bug and inconsistent tied-weight serialization are related only at a very broad theme level." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Template compilation error and GraniteMoEHybrid invalid-method call are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate bug and MIMI padding-mask bug are different modalities and code paths." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "SigLIP2 nonsensical outputs and OmDet-Turbo image-size mismatch are unrelated." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Custom model initialization regression and non-persistent buffer serialization are different failures." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Llava Next segfault and Qwen3OmniMoe video ValueError are different multimodal loading/processing bugs." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Notebook custom-model init breakage and non-persistent buffer serialization are not the same bug." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "These are the same custom-model initialization regression, with the notebook case just being a more specific manifestation." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Same v5 non-persistent-buffer junk regression; the titles are essentially 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": 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-10000-400", + "nodes": [ + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43296", + "issue:43377", + "issue:43404", + "issue:43421", + "issue:43425", + "issue:43531", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43742", + "issue:43792", + "issue:43818", + "issue:43819", + "issue:44062", + "issue:44079", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44479", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44811", + "issue:44877", + "issue:44987", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45084", + "issue:45127", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45542", + "issue:45663" + ], + "soft_pairs": [ + "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:43632|issue:44756", + "issue:42915|issue:44291", + "issue:44062|issue:44987", + "issue:44479|issue:45325", + "issue:36032|issue:43742", + "issue:43296|issue:43531", + "issue:40444|issue:45325", + "issue:42915|issue:44589", + "issue:43611|issue:44534", + "issue:43531|issue:44877", + "issue:36032|issue:41720", + "issue:43425|issue:44403", + "issue:44079|issue:45663", + "issue:40990|issue:44811", + "issue:41553|issue:43582", + "issue:43632|issue:44534", + "issue:42175|issue:45542", + "issue:43421|issue:44568", + "issue:43818|issue:45357", + "issue:43792|issue:44589", + "issue:44403|issue:44991", + "issue:44683|issue:45020" + ], + "prepared_review_unit_hash": "ab691ecabb728ea9eab336aea30a7cc55eb46fb8abc104b3459cfe742584975c", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36439, + "estimated_input_tokens": 9110, + "estimated_eval_tokens": 18476 + }, + "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 regressions, with a noticeable concentration around v5 model/tokenizer loading behavior. None of the soft pairs is close enough to be treated as a true duplicate or merge-worthy PR-equivalent.", + "confidence": 0.88, + "canonical_issue_reason": "Issue 43632 is the most central representative of the dominant theme: a core Transformers v5 load-path regression involving internal initialization state, which plausibly underlies several later loading failures.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43632 is the clearest cluster anchor because it describes a broad v5 regression in model loading, whereas most other issues are narrower, model-specific, or only loosely related.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Different bugs: noisy load-time logging vs a tokenizer regex crash." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Unrelated code paths: MIMI padding-mask batching bug vs DAC latent reconstruction mismatch." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Different failures: Qwen3Moe FP8 config handling vs template-node compilation error." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "Different problems in different subsystems: torch.export/compilable checks vs image-size processor mismatch." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Both are load failures, but for different models and different root causes." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": false, + "reason": "Same model family, but one is a saved-checkpoint regression and the other is incorrect visual-encoder key saving; not the same concrete bug." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Different model-loading failures with no shared root cause." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Different subsystems: fbank feature extraction vs compiled flex_attention on newer torch." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both touch weight tying, but one is a specific missing lm_head tie and the other is LoRA merge collapse after vocab extension." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "Unrelated: internal v5 initialization flag break vs mmap OOM on Strix Halo." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Different root causes: Qwen3Moe FP8 config failure vs init_empty_weights argument compatibility." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer AddedToken argument error is unrelated to the physical-intelligence/fast load regression." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Both are Qwen VL regressions, but one is broad video-input behavior and the other is a specific still-image rope-index scaling bug." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Different failures: T5 tokenizer method conflict vs MobileLLM load key error." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "Different model-specific failures: PaddleOCR-VL loading in vLLM vs Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "Different bugs: iterable dataset with multiple images vs temporal position scaling for still images." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "FP8 config failure and missing Float8 storage are related only at a very high level." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Different Transformers v5 regressions: base_model_prefix loading vs junk-filled non-persistent buffers." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Both mention config/loading issues, but they affect different models and different mechanisms." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer-loading conflict vs CUDA assert during auto device mapping are unrelated." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch version incompatibility is unrelated to a noisy load-message issue." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment bug and Gemma4 FSDP2 shared_kv_states bug are different code paths." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Perplexity regression on GPT-OSS is unrelated to Whisper batch_decode special-token handling." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Bad tokenizer error messaging and Apple Silicon allocator TypeError are unrelated." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "Different v5 regressions: initialization flag handling vs buffer initialization contents." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both concern TensorFlow packaging, but one is missing TF in the torch extra and the other is tensorboard-only causing an undefined backend error." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Both involve special tokens, but one is runtime post-processor syncing and the other is add_special_tokens not inserting BOS/EOS." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Different bugs: Video-LLaVA tower architecture/weights vs Qwen3.5 save_pretrained key serialization." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure and missing Float8 storage are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic load noise and a tokenizer load failure for a specific model are not the same bug." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Compiled flex_attention failure on torch>=2.9 is unrelated to remote_code loading regressions." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43381", + "issue:43454", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43646", + "issue:43701", + "issue:43749", + "issue:43792", + "issue:43856", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44387", + "issue:44462", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44589", + "issue:44617", + "issue:44829", + "issue:44898", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45800", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45923" + ], + "soft_pairs": [ + "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:43482|issue:43525", + "issue:44291|issue:44589", + "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:44265|issue:45072", + "issue:43632|issue:45072", + "issue:45325|issue:45381", + "issue:43994|issue:44898", + "issue:43611|issue:43646", + "issue:42175|issue:43296", + "issue:43454|issue:44466", + "issue:45325|issue:45834", + "issue:43792|issue:43957", + "issue:42915|issue:43482", + "issue:43296|issue:45800", + "issue:44387|issue:44617", + "issue:43701|issue:45325", + "issue:44479|issue:45823" + ], + "prepared_review_unit_hash": "46d4cd011c4b00cffd0bd48c91c8946f1cb9ec6a129c8d57f5e79d412a00318d", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36752, + "estimated_input_tokens": 9188, + "estimated_eval_tokens": 18632 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly distinct Transformers regressions across different models/subsystems. A small number of pairs look close enough to be duplicates, especially the v5 model-initialization/loading reports and the Qwen2.5-VL position-id regressions.", + "confidence": 0.64, + "canonical_issue_reason": "issue:43611 is the broadest and clearest representative of the v5 model-loading / initialization regression class, with a generic scope that can absorb closely related reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the best anchor issue for this cluster because it describes a concrete, widely applicable Transformers v5 breakage rather than a narrow model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but the symptoms differ too much: one is repository tokenizer.json lookup, the other is Granite-only token-id corruption." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Different multimodal training failures: multi-GPU training vs iterable-dataset multi-image prompting." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both mention degenerate training, but one is an attention backend bug and the other is a classification-loss setup bug." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Different problems: device_map/offload loading failure vs MoE training memory inefficiency." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated tokenization/model-loading issue versus a CLIPTokenizer max-length constant bug." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "One is an eval-mode checkpointing restriction; the other is a missing model capability flag on trust_remote_code." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Different generation/training scenarios and different failure modes." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment bug is unrelated to non-persistent buffer serialization junk." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Both are loading-related, but one is GGUF/v5 loading and the other is a missing config attribute on Llama4." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Different loading-time failures: unexpected init argument vs missing Float8 storage class." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Both involve batching/padding, but they affect different models and different code paths." + }, + { + "left": "issue:43749", + "right": "issue:44079", + "accept": false, + "reason": "FSDP CPU RAM efficient loading is unrelated to ModelOutput key handling." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both concern Qwen2.5-VL/video-image inputs, but the reported bugs are different enough to keep separate." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Non-persistent buffer filling is not the same issue as ModelOutput key assignment." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Completely different failures: CUDA assert under device mapping vs tokenizer regex patch crash." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Different regressions in Qwen3-MoE sliding-window behavior versus init_empty_weights argument handling." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export compatibility and dtype mismatches are different concrete bugs." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "_is_hf_initialized flag breakage is unrelated to bfloat16 dtype mismatches." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": true, + "reason": "Both point to the same Qwen2.5-VL 5.3.0 position-id / rope-index generation regression, just surfaced through still-image vs video paths." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different vision-model bugs: SigLIP2 tokenization/output issue vs Perceiver resolution/interpolation failure." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": true, + "reason": "Both describe Transformers 5 breaking custom-model loading/initialization on the same code path, and could plausibly be fixed together." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "Packaging/backend inclusion and model-loading failures are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "One is tied lm_head weights causing bad generation; the other is serialization inconsistency of lm_head.weight across devices." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Position-id scaling bug and Kosmos2.5 long-OCR index error are different multimodal issues." + }, + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "Different model-loading/runtime failures with no clear shared mechanism." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Qwen3MoE FP8 config failure is unrelated to Qwen2.5 GGUF loading under v5." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "PaddleOCR-VL loading failure and torch/transformers version incompatibility are different root causes." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Int4 reserved-memory regression and Sam3Video OOM are not the same bug." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint-resume key mismatch is unrelated to Qwen2.5-VL position-id generation." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Video-input regression in Qwen-family models is unrelated to Gemma4 device mismatch during forward." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet: issue:43611 is a reasonable broad anchor for the v5 model-loading/init regression class, and the two accepted soft edges are conservative enough based on their titles. The false soft-edge calls also stay appropriately narrow and do not overclaim duplicate status." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42947", + "issue:43257", + "issue:43262", + "issue:43296", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43454", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43720", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:44008", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44466", + "issue:44488", + "issue:44534", + "issue:44568", + "issue:44610", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45357", + "issue:45584", + "issue:45701", + "issue:45759", + "issue:45823", + "issue:45920", + "issue:45923" + ], + "soft_pairs": [ + "issue:43881|issue:44488", + "issue:43606|issue:44265", + "issue:44291|issue:44877", + "issue:44568|issue:45920", + "issue:43262|issue:43746", + "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", + "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:43404|issue:45357" + ], + "prepared_review_unit_hash": "051f45a1991f839b32b3d94781da8de31ce7612c59c3f65f90893c54ae915878", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36705, + "estimated_input_tokens": 9177, + "estimated_eval_tokens": 18610 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated regressions spanning loading, tokenization, multimodal processing, and training/runtime bugs. None of the soft pairs look like true duplicates; the closest recurring theme is offloading/device_map/quantization loading behavior, where issue 43873 is the broadest representative.", + "confidence": 0.63, + "canonical_issue_reason": "Issue 43873 is the broadest and most representative of the recurring offloading/quantization-loading failures in this set, with active discussion and a general symptom description.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is the best global representative because it captures the broadest underlying failure mode among the candidates and is currently open/actively relevant.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Different failures in different models: glm-4v-9b load failure vs tokenizer/model loading for cjvt/sleng-bert." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch in Bark is unrelated to torch.export failures from torch_compilable_check." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "Unexpected _is_hf_initialized arg during init_empty_weights is a different bug from strict granite_speech config loading." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "One is add_special_tokens/BOS-EOS behavior for a specific tokenizer; the other is wrong token IDs across several models." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting and GraniteSpeech PEFT local-checkpoint loading are separate code paths." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "High perplexity on gpt-oss-20b is unrelated to Whisper empty-transcription failure after align_special_tokens." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "Bark voice_preset not working is not the same as progressive generation with inputs_embeds and past_key_values." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Offloading with quantization is unrelated to tokenizer changes across transformers versions." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "LayoutLMv2Tokenizer padding/NER crash is a tokenizer bug, not a torch.export/torch_compilable_check issue." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Different runtime problems: generation with inputs_embeds/past_key_values versus Qwen3 device-map cudaErrorAssert." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Both mention model loading behavior, but Bark CPU-offload mismatch and DAC.from_latents/STE mismatch are distinct bugs." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both involve device_map=auto, but they affect different models and failure modes (cuda assert vs forward device mismatch)." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "lm_head weight tying in AyaVision is unrelated to Bamba torch.compile SDPA failures." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI padding-mask inconsistency and OmDet-Turbo processor/model resolution mismatch are unrelated." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Dequantized-model loading with gpu+cpu offload is related in theme but not the same concrete bug as offloading with quantization." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 docs/example errors and Qwen2.5-VL iterable-dataset multi-image finetuning failures are different problems." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to not unwrapping text_config for multimodal Qwen configs." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "AyaVision tied-weights bug is unrelated to dtype mismatches in bfloat16 inference." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "MIMI batched-vs-single output mismatch from padding masks is unrelated to Perceiver interpolation failure." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is unrelated to a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "BEiT reduce_labels returning one label is unrelated to Molmo missing all_tied_weights_keys." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "Perplexity regression and gradient-checkpointing inefficacy are different training/evaluation issues." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "lm_head serialization inconsistency by device is distinct from Qwen3.5 visual-encoder key serialization." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Model loading failure for glm-4v-9b is unrelated to tokenizer loading for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and non-persistent buffer junk are separate regressions." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "AyaVision weight tying and BitNet packed-weight unpacking are different loading/serialization bugs." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3 device-map cuda assert and Llava Next segfault are different model-specific failures." + }, + { + "left": "issue:43296", + "right": "issue:44877", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM is unrelated to strict granite_speech config loading." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug is not the same as Qwen3 auto device mapping cuda assert." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Gemma 3n variable-name collision in forward is unrelated to Gemma4 PLE device mismatch." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Mistral3 lm_head tying and Qwen3.5 visual encoder key serialization are separate 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": 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-10000-400", + "nodes": [ + "issue:36010", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43316", + "issue:43317", + "issue:43322", + "issue:43421", + "issue:43450", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43708", + "issue:43720", + "issue:43782", + "issue:43873", + "issue:43881", + "issue:44186", + "issue:44361", + "issue:44387", + "issue:44423", + "issue:44479", + "issue:44610", + "issue:44671", + "issue:44743", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:44918", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45161", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45812" + ], + "soft_pairs": [ + "issue:45632|issue:45684", + "issue:42617|issue:42915", + "issue:40990|issue:45701", + "issue:44186|issue:45072", + "issue:44423|issue:45406", + "issue:43577|issue:43708", + "issue:44387|issue:45704", + "issue:43873|issue:44756", + "issue:44877|issue:44991", + "issue:36331|issue:42617", + "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", + "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" + ], + "prepared_review_unit_hash": "1d9946bb5e5a04a38c68fc0fa4f4f71823399f10b4fb9e95eb04324b7718722b", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36462, + "estimated_input_tokens": 9116, + "estimated_eval_tokens": 18488 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: the soft pairs share broad surface themes like tokenizers, loading, or offloading, but they do not look like the same underlying bug. I rejected all soft edges and picked the most connected issue only as a weak triage anchor, not as a true duplicate canonical.", + "confidence": 0.47, + "canonical_issue_reason": "issue:42617 is the most connected node in the soft-link graph, so it is the best weak anchor for triage, but its neighbors point to different failure modes and it is not a true duplicate center.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42617 is the most central issue among this noisy set; it is the best representative for manual review, though not a safe duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Cache-path collisions for trust_remote_code and read-only permission propagation are different bugs." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "3d_parallel.py execution failure and Qwen3Moe FP8 loading failure are unrelated code paths." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "High perplexity on WikiText-2 and a tokenization regression are not the same failure." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes and bf16 dtype mismatches have different root causes." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "A continuous-batching multimodal crash with str.to and a missing _tokenizer on Gemma4Processor are distinct bugs." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Blip2 dtype handling and Trainer resume max_steps calculation are unrelated." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Int4 quantization memory growth and an apex RMSNorm memory leak are different issues." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Generic quantization offloading problems and Strix Halo mmap OOM are not the same bug." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Strict config loading and tokenizer loading failure are separate code paths." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "CustomTrainer.compute_loss signature breakage is unrelated to 3d_parallel.py startup failure." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "torch.compile SDPA failure in Bamba and packed-weight unpacking in BitNet are different." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Trainer compute_loss API breakage and Qwen3Moe FP8 loading failure do not match." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Batched video shape handling and version-dependent tokenization changes are not the same bug." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "OmDet-Turbo input-size mismatch and Perceiver non-default-resolution failure are similar only at a high level, not the same code path." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer crashes in LayoutLMv2 and image-size mismatch in OmDet-Turbo are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44918", + "accept": false, + "reason": "Qwen video-input regression and Qwen3.5 SFT embedding unpacking are different failures." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "SDPA/compile failure and CPU-offload device mismatch are separate issues." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and 3d_parallel.py execution failure are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Gemma3TextConfig API mismatch and GPTNeoX rotary_pct reload bug are different config problems." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Qwen3Moe FP8 loading and dequantized model offload failures are different quantization/loading paths." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Processor.tokenizer regression and tokenization changes are not the same underlying bug." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and glm-4v-9b loading failure are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "CamemBERT masked-LM regression and Qwen3_5 recurrent-state reset are distinct model bugs." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Auto device mapping failure and GPT-OSS tensor-parallel failure are different model-parallelism issues." + }, + { + "left": "issue:45081", + "right": "issue:45812", + "accept": false, + "reason": "Mistral regex patch crash and Granite token-id regression are both tokenizer bugs but not the same fix." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and StableLm pad_token_idx omission are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Video-processing ValueError in Qwen3OmniMoe and weight_only loading error in Qwen3VL are different multimodal failures." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "GenerationMixin import error and 3d_parallel.py execution failure have no common code path." + }, + { + "left": "issue:43421", + "right": "issue:45812", + "accept": false, + "reason": "Runtime post-processor updates for special tokens and Granite token-id regression are different tokenizer behaviors." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "LayoutLMv2Tokenizer NER/padding crash and MLukeTokenizer task AttributeError are separate tokenizer bugs." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3Moe FP8 failure and Llava Next segfault are unrelated model-loading problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:41720", + "issue:42490", + "issue:43329", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43550", + "issue:43606", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:43927", + "issue:44062", + "issue:44077", + "issue:44368", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44514", + "issue:44545", + "issue:44589", + "issue:44625", + "issue:44898", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45059", + "issue:45072", + "issue:45357", + "issue:45381", + "issue:45406", + "issue:45464", + "issue:45479", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:45923" + ], + "soft_pairs": [ + "issue:43388|issue:45059", + "issue:45406|issue:45464", + "issue:44492|issue:44509", + "issue:33357|issue:35707", + "issue:43421|issue:45701", + "issue:43606|issue:43720", + "issue:41720|issue:43746", + "issue:30064|issue:34634", + "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", + "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:44918|issue:44936" + ], + "prepared_review_unit_hash": "ab08afc26cf4eb06793ff123305a7fe25bd97bc0b7807091c4826bf10e667441", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 36080, + "estimated_input_tokens": 9020, + "estimated_eval_tokens": 18296 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous. Only one soft pair is a clear duplicate: the Qwen2_5_VLProcessor.apply_chat_template crash on batched input with padding=False.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 44545 is the best canonical issue because it is an exact duplicate of 44514, has the clearer later report, and directly names the same Qwen2_5_VLProcessor padding=False crash.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44545 is the strongest issue candidate overall in this cluster because it is the only clear duplicate target and most precisely represents the shared bug in the accepted pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Different problems and code paths: metric gathering drops tuple labels vs SAM3 text/bbox behavior." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Different model/runtime failures: Gemma4Processor missing _tokenizer vs Qwen3.5 chat streaming API failure." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "One is a typo in cache strategy docs, the other is stale pipeline docs; related documentation cleanup but not the same bug." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "Unrelated failures: MacOS bus error with a CLIP model vs progressive generation with inputs_embeds/past_key_values." + }, + { + "left": "issue:43421", + "right": "issue:45701", + "accept": false, + "reason": "Runtime special-token post-processor update bug is different from a broad tokenization regression report." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "Both involve loading/offload, but they are different concrete failures in different models: CPU offload device mismatch vs packed-weight unpacking." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Different subsystems and failures: Qwen3 auto device mapping cuda assert vs GraniteSpeech PEFT local checkpoint loading." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Different processors and failures: void segmentation maps vs Bark voice_preset behavior." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "Different areas entirely: torch.compile/SDPA issue vs tokenizer special-token registration." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization/offloading behavior is unrelated to a T5 apex FusedRMSNorm memory leak." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same exact Qwen2_5_VLProcessor.apply_chat_template batched-input crash with padding=False; clearly the same underlying bug." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Both mention classification settings, but one is config propagation and the other is degenerate loss computation." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Different failure modes and models; one is bark-small CPU offload mismatch, the other is a tokenizer special-token bug." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is not the same as unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "A trailing-slash import warning is unrelated to undefined helpers in multimodal token counting." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "Different products and code paths: BarkProcessor voice_preset vs Qwen3 auto device mapping cudaErrorAssert." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token accounting is unrelated to progressive generation with embeddings and cache." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Both involve special-token handling, but one is runtime post-processor refresh and the other is config save/load losing custom token IDs." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "A warning about tie_word_embeddings is unrelated to incorrect visual encoder keys in save_pretrained." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen VL bugs: from_pretrained weight_only handling vs incorrect video vision_position_ids." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Different subsystems: Perceiver image resolution interpolation vs dtype mismatches in SwitchTransformers/TimmWrapperModel inference." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are model-loading/tokenizer issues, but they involve different models and different concrete tokenizer failures." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "PatchTSMixer post_init policy and Qwen3.5 num_labels propagation are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Different loading failures: a specific model regression vs PIL backend image processors requiring torchvision." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Different Qwen 3.x issues: config unwrapping vs CPU-offload incompatibility in rotary embedding." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to missing Float8 storage support." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "AddedToken argument duplication is a tokenizer construction bug, not the same as a model load failure for ScandiBERT." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "A T5 apex RMSNorm memory leak is unrelated to a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Both are trainer-related, but one is unpacking Qwen3.5 input embeddings in TRL SFT and the other is evaluate() after train() failing." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the one accepted pair (44514/44545) is supported by the identical issue title and matching crash description. The other soft-edge verdicts are conservative and avoid overstating similarity." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:40444", + "issue:42886", + "issue:43295", + "issue:43322", + "issue:43335", + "issue:43381", + "issue:43421", + "issue:43525", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43596", + "issue:43618", + "issue:43653", + "issue:43697", + "issue:43716", + "issue:43720", + "issue:43746", + "issue:43756", + "issue:43819", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44008", + "issue:44186", + "issue:44220", + "issue:44610", + "issue:44661", + "issue:44671", + "issue:44843", + "issue:45072", + "issue:45081", + "issue:45183", + "issue:45356", + "issue:45381", + "issue:45479", + "issue:45588", + "issue:45663", + "issue:45685", + "issue:45704", + "issue:45759", + "issue:45834", + "issue:45859", + "issue:45907", + "issue:45923" + ], + "soft_pairs": [ + "issue:39692|issue:44610", + "issue:43421|issue:43653", + "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:39692|issue:43873", + "issue:45356|issue:45907", + "issue:43746|issue:43873", + "issue:43596|issue:44220", + "issue:40444|issue:43746", + "issue:43746|issue:45381", + "issue:43540|issue:44008", + "issue:43720|issue:43819", + "issue:44220|issue:44661", + "issue:45183|issue:45479", + "issue:43618|issue:43697", + "issue:30064|issue:36010", + "issue:43653|issue:43720", + "issue:43335|issue:43756", + "issue:43716|issue:44008", + "issue:43295|issue:45834", + "issue:43550|issue:45072", + "issue:43550|issue:43819", + "issue:43322|issue:45685" + ], + "prepared_review_unit_hash": "40a6c2843afed04e8fc0fba671695515574bc815e6475e0a95cbaea95a79152e", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37181, + "estimated_input_tokens": 9296, + "estimated_eval_tokens": 18848 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is largely heterogeneous; most soft pairs are separate bugs touching different models or code paths. The only convincing duplicate pair is the HF_HUB_OFFLINE / tokenizer cache failure around `_patch_mistral_regex` (`issue:42886` and `issue:44843`).", + "confidence": 0.9, + "canonical_issue_reason": "For the only clear duplicate subcluster, `issue:44843` is the better canonical because it identifies the concrete root cause (`_patch_mistral_regex` unconditionally calling `model_info()` in offline mode) rather than only the broader offline-cache symptom.", + "canonical_pr_reason": null, + "best_issue_reason": "`issue:44843` is the clearest representative of the one true duplicate pair in this set; the rest of the issues are too unrelated to share a single canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Both are vision-model example failures, but one is a SigLIP2 doc/example mismatch and the other is an OmDet-Turbo processor/model size mismatch; not the same bug." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Both involve tokenizers and special tokens, but one is about runtime post-processor updates and the other about BigBird mask-token registration/decoding; different defects." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Different problems: a Gemma4 FSDP2 per-layer `shared_kv_states` rebuild bug versus a missing `supports_gradient_checkpointing` flag on a trust_remote_code variant." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "One is a flash-attention crash on `s_aux=None`; the other is a composite Qwen config unwrap issue in `from_config`. No shared concrete code-path bug." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": true, + "reason": "Same underlying offline-tokenizer failure: both report HF_HUB_OFFLINE/cache-loading breakage, and `issue:44843` identifies the concrete `_patch_mistral_regex`/`model_info()` cause." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Both mention training/runtime behavior, but one is gradient checkpointing in eval mode and the other is gradient explosion with ZeRO-3; not duplicates." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Tokenizer-related, but different failure modes: post-processor not updating special tokens versus LayoutLMv2 NER/batched padding crashes." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different model regressions with different symptoms and code paths; no evidence they are the same bug." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "One is a missing `pad_token_id` attribute on Llama4Config; the other is a Qwen3-MoE `sliding_window` issue. Distinct." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "T5/apex RMSNorm memory leak and Qwen3_5Moe CPU-offload incompatibility are unrelated." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both touch Mistral-related tokenizer logic, but one is a hard crash in `_patch_mistral_regex` and the other is a codec-handling/regression report with a warning issue; not the same concrete bug." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "A doc/example mismatch is not the same as a quantization/offloading behavior bug." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Kimi tokenizer regression and `list_repo_files` not forwarding `revision` are unrelated." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "PEFT/local checkpoint loading for GraniteSpeech and quantization offloading behavior are different problems." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "Deepspeed ZeRO-3 initialization with BertModel and `_torch_extract_fbank_features()` are unrelated code paths." + }, + { + "left": "issue:40444", + "right": "issue:43746", + "accept": false, + "reason": "Qwen2.5-VL iterable dataset/multi-image finetuning and GraniteSpeech PEFT checkpoint loading are different bugs." + }, + { + "left": "issue:43746", + "right": "issue:45381", + "accept": false, + "reason": "PEFT/local checkpoint loading and Qwen2.5-VL video `vision_position_ids` are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Qwen3OmniMoe video-processing ValueError and Gemma 3n `audio_mel_mask` name-collision AttributeError are different defects." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "BitNet accelerate-loading unpacking issue and DAC `from_latents`/STE mismatch are unrelated." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "An audio feature-extraction issue and an `add-new-model-like` tokenizer-mapping failure are not the same bug." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Wrong `input_ids`/`inputs_embeds` validation messaging is separate from the single-label-classification zero-loss bug." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "CLIP attentions not assigned and RTDetrV2 output differences in v5 are different model regressions." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Image processor void-segmentation-map failure and the `GenerationMixin` import error are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "BigBird special-token registration and BitNet packed-weight loading are entirely different issues." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation and Smollm3 RoPE-layer dropping are unrelated model-config bugs." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Mistral-3 image preprocessor dtype mismatch and Gemma 3n audio-mask name collision are different problems." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "A processor/tokenizer regression and a Kosmos2.5 long-OCR index error do not share the same underlying cause." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure and dtype mismatches in SwitchTransformers/TimmWrapperModel are distinct." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Bamba SDPA/torch.compile issue and DAC latent conversion mismatch are unrelated." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Llava Next segfault and MPS histogram int support error are not the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analyst stays conservative and grounds the summary in the packet. The single accepted pair, issue:42886 and issue:44843, is plausibly the same offline-tokenizer / `_patch_mistral_regex` failure, and the other soft pairs are reasonably rejected as distinct bugs rather than over-merged." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:30064", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43643", + "issue:43644", + "issue:43653", + "issue:43749", + "issue:43782", + "issue:43818", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43994", + "issue:44062", + "issue:44186", + "issue:44442", + "issue:44448", + "issue:44479", + "issue:44485", + "issue:44488", + "issue:44509", + "issue:44561", + "issue:44671", + "issue:44805", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45381", + "issue:45584", + "issue:45701", + "issue:45834" + ], + "soft_pairs": [ + "issue:44062|issue:44488", + "issue:44186|issue:44898", + "issue:44843|issue:45081", + "issue:43531|issue:43643", + "issue:43653|issue:43819", + "issue:43577|issue:43818", + "issue:43643|issue:44877", + "issue:43540|issue:45381", + "issue:43421|issue:44811", + "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:44485|issue:44509", + "issue:44821|issue:45084", + "issue:30064|issue:36331", + "issue:41720|issue:44991", + "issue:43644|issue:43749", + "issue:44186|issue:44442", + "issue:43825|issue:43827", + "issue:44479|issue:45834", + "issue:43577|issue:43582", + "issue:44805|issue:44936", + "issue:43606|issue:45072", + "issue:44671|issue:45701" + ], + "prepared_review_unit_hash": "302a991f7d2001395c60bed868515b7b351c9731cb7178af2811b38e368edab4", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35265, + "estimated_input_tokens": 8817, + "estimated_eval_tokens": 17890 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated bug reports that only share broad themes like tokenizer, multimodal, and v5 regression language. The soft pairs are mostly superficial similarities and do not look like the same underlying bug or change.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45020 is the broadest open umbrella here, covering recent version regressions around remote_code/config loading; the other issues are narrower, model-specific failures.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the best anchor for the cluster because it is the most general and central regression report, even though it does not duplicate the other issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "Different failure modes: AddedToken constructor error vs tokenizer/model loading failure for a specific repo." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Different models and code paths: LayoutLMv2 tokenizer padding/NER crash vs Perceiver image classification resolution/interpolation bug." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both touch _patch_mistral_regex, but one is an offline model_info call and the other is an AttributeError on tokenizer internals; not the same bug." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Unrelated areas: Qwen3-MoE sliding_window behavior vs AutoConfig trust_remote_code field loss." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Tokenizer special-token registration bug vs DAC latents/STE forward mismatch; no common code-path." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Different multimodal models and defects: dtype propagation in Blip2 vs missing temporal attention / shared weights in Video-LLaVA." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both involve config loading, but one is missing fields from trust_remote_code and the other is strict config rejection for granite_speech." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different Qwen video bugs: Qwen3OmniMoe video processing ValueError vs Qwen2.5-VL vision_position_ids regression." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Runtime special-token/post-processor mutation issue vs Whisper batch_decode skip_special_tokens handling; different components." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Loading error with weight_only=True vs generation failure with flash-attention; not the same concrete bug." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are v5 regressions, but one is Pegasus output drift and the other is CamemBERT masked-LM prediction corruption; different models and causes." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty transcription generation vs Kosmos2.5 long OCR index error; different models and failure signatures." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "Docs/error-message wording around removed pipeline tasks vs a SigLIP2 model/pipeline output problem; not the same bug." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Related only at a broad level: 44561 is a specific remote_code break from removing is_torch_fx_available, while 45020 is a generic umbrella report." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 docs/example errors are unrelated to Kosmos2.5 OCR index failures." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion vs outdated docs about removed pipeline tasks; different issues." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading vs Jinja/template compilation TypeError; unrelated subsystems." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image processor segmentation-map failure vs trainer.compute_loss signature mismatch; no overlap in code path." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Qwen3 auto device-map CUDA assert vs tokenizer loading regression for EMBEDDIA/est-roberta; different problems." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Non-persistent buffer initialization bug vs FSDP CPU RAM efficient loading breakage; related only by version." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Tokenizer crash in LayoutLMv2 vs AutoTokenizer loading failure for FastSpeech2Conformer; different artifacts and code paths." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "Both reference removed v5 pipeline tasks, but one is a runtime message issue and the other is stale docs; not the same bug/change." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Qwen video-input regression vs Kosmos2.5 OCR long-input index error; unrelated model families." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Blip2 dtype propagation issue vs Apple Silicon caching_allocator_warmup TypeError; distinct failures." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Tensor mask shape mismatch vs trainer.evaluate after train lifecycle bug; different concrete problems." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "CPU offload device mismatch in bark-small vs bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT v5 prediction regression vs general tokenization changes across versions; 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": 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-10000-400", + "nodes": [ + "issue:34634", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:42175", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43452", + "issue:43526", + "issue:43582", + "issue:43638", + "issue:43650", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43746", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43854", + "issue:43906", + "issue:43927", + "issue:44060", + "issue:44291", + "issue:44368", + "issue:44554", + "issue:44589", + "issue:44610", + "issue:44811", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45042", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45406", + "issue:45459", + "issue:45538", + "issue:45600", + "issue:45803" + ], + "soft_pairs": [ + "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:43329|issue:45600", + "issue:43742|issue:45020", + "issue:43653|issue:44912", + "issue:43638|issue:44936", + "issue:43746|issue:44811", + "issue:43452|issue:44991", + "issue:38175|issue:44610", + "issue:44589|issue:45042", + "issue:43854|issue:44977", + "issue:43819|issue:43927", + "issue:39692|issue:44912", + "issue:43257|issue:43329" + ], + "prepared_review_unit_hash": "3cfa70e286fe3ffcd2c7c02b2e87b8a7628b57cd9fe1d17b595c5a62bf7a38fb", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36472, + "estimated_input_tokens": 9118, + "estimated_eval_tokens": 18492 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated Transformers bug reports spanning model loading, tokenizers, processors, quantization, and training. None of the soft pairs look like the same underlying defect, so all candidate merges are rejected. The broadest representative issue is the remote_code regression report.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45020 is the broadest umbrella report about recent Transformers versions breaking remote_code models, so it best represents this mixed regression cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the most general and centrally scoped issue in the set; the others are narrower, model-specific failures rather than a shared duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "Different subsystems and failure modes: BarkProcessor voice_preset vs GenerationMixin import regression." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "Different bug classes: processor image size mismatch vs dtype mismatch in inference." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Both are model-loading related, but one is weight_only loading and the other is a rope-index temporal-position bug." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "BitNet packed-weight loading failure is unrelated to bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Packaging/backend dependency issue vs tokenizer regex patch crash; not the same defect." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Both involve tokenizer/processor code, but different objects and crashes in different call paths." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "Both concern loading flows, but one is FSDP CPU RAM-efficient loading and the other is init_empty_weights argument handling." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "Generic placeholder issue vs isolated reproduction of a specific external bug; not duplicates." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Different tied-weights warnings affecting different model families and configurations." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon allocator TypeError is unrelated to CUDA AMP crash in a detection loss." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio sampling-rate default bug is unrelated to DAC latent/forward-pass mismatch." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BEiT reduce_labels bug and RTDetrV2 output drift are separate model-specific defects." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "No shared code path: BEiT label reduction vs MPS attention correctness on differing head dims." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve error handling, but one is protobuf decode error masking and the other is a bare except in tensor conversion." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "Unrelated: BarkProcessor preset behavior vs CustomTrainer compute_loss signature change." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Template compilation error and CLIPTokenizer max length bug are unrelated." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Multimodal token counting bug is unrelated to stale auto_mappings config references." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "Specific MobileLLM key error does not match the broader remote_code regression." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Tokenizer special-token registration bug is unrelated to MXFP4 quantization fallback." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Zero3/non-pretrained Bert index error is not the same as trainer.evaluate() failing after train()." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "Local PEFT checkpoint loading issue is unrelated to Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "gguf_file support failure is different from tokenizer loading regression for est-roberta." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 are not the same as OmDet processor input-size mismatch." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Storage lookup failure for float8 is unrelated to torchvision dependency requirements in PIL image processors." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Model loading/test issue for GLM-4.7-Flash is not the same as flash-attention generation failure in Qwen3.5." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC latent mismatch and DiaConfig custom token ID loss are separate generation/model-config bugs." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "SigLIP2 doc example errors do not match MXFP4 quantization fallback behavior." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MOE weight conversion/loading issue is unrelated to multimodal token counting undefined variables." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:33357", + "issue:34567", + "issue:36010", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43454", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43643", + "issue:43645", + "issue:43696", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43931", + "issue:44291", + "issue:44479", + "issue:44568", + "issue:44610", + "issue:44756", + "issue:44811", + "issue:44849", + "issue:44912", + "issue:44987", + "issue:45042", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45488", + "issue:45538", + "issue:45701", + "issue:45759", + "issue:45859", + "issue:45907" + ], + "soft_pairs": [ + "issue:36010|issue:43645", + "issue:43720|issue:44610", + "issue:43550|issue:44912", + "issue:43454|issue:43577", + "issue:43257|issue:43696", + "issue:43262|issue:44811", + "issue:43329|issue:44479", + "issue:42490|issue:43257", + "issue:43531|issue:45042", + "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:44849|issue:45701", + "issue:43257|issue:43931", + "issue:43606|issue:45859", + "issue:43697|issue:45357", + "issue:43257|issue:43756", + "issue:44479|issue:45381", + "issue:43452|issue:43582", + "issue:44987|issue:45538", + "issue:41720|issue:43540", + "issue:44291|issue:45042", + "issue:42915|issue:43742", + "issue:44987|issue:45907", + "issue:33357|issue:34567" + ], + "prepared_review_unit_hash": "3282751fd1ac6e24f64b39d934f4103215310d01de41d429960799e10ff376e0", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36597, + "estimated_input_tokens": 9150, + "estimated_eval_tokens": 18556 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: tokenization regressions, model-loading bugs, multimodal/video processor issues, and training API breakages. None of the soft pairs look like the same underlying bug/change, so I would not deduplicate them.", + "confidence": 0.94, + "canonical_issue_reason": "issue:45701 is the broadest open issue and best fits an umbrella for the tokenizer/version-regression subset, with multiple inbound references and a title that captures the general symptom pattern.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45701 is the most central and reusable issue in this set; it\u2019s broad enough to represent the shared tokenizer-regression theme better than the highly specific model- or API-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "Different breakages: a missing GenerationMixin import vs Jupyter-defined custom model initialization in v5." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "Both are loading/vision-adjacent failures, but one is about unpacking packed BitNet weights during accelerate loading and the other about processor image size mismatch." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "Different concrete problems: torch.compile with SDPA on Bamba vs MXFP4 quantization fallback on gpt-oss-20b." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "AyaVision weight tying bug is unrelated to Blip2 dtype propagation on model load." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MOE weights not converted during accelerate+deepspeed is distinct from a plain CUDA OOM on gpt-oss-20b." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio apply_chat_template sampling-rate default bug is unrelated to Whisper batch_decode skipping special tokens." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Undefined video token counting helpers vs a broader video-input regression in several VLMs are related by modality only, not the same bug." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Local-path import warning with trailing slash is unrelated to Qwen3 MOE weight conversion under accelerate/deepspeed." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Sliding window behavior in Qwen3-MoE is not the same as PIL-backed image processors wrongly requiring torchvision." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "ImportError for GenerationMixin and unexpected num_items_in_batch in CustomTrainer are separate API regressions." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both tokenizer-related, but one is add_special_tokens/BOS-EOS behavior while the other is codec handling and warning text in Kimi-K2.5." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and disabling mmap on Strix Halo are different memory-loading issues with different triggers." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Missing fields from trust_remote_code AutoConfig is not the same as AutoModelForCausalLM.from_config failing to unwrap text_config in composite Qwen configs." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Different tokenizer regressions: add_special_tokens handling vs LlamaTokenizer overriding ByteLevel pre-tokenizer." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "TrainerState token counting not updating is unrelated to high perplexity on a specific model/dataset." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Undefined helper in multimodal token counting vs batched video processor shape mismatch are separate video-path bugs." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 output_hidden_states bug is model-specific and not the same as the broad tokenization-version regression." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3 MOE accelerate/deepspeed conversion failure is different from Qwen3-VL weight shape mismatch on model load." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "CPU offload device mismatch in bark-small is unrelated to Qwen3_5MoeTextRotaryEmbedding CPU offload compatibility." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetrV2 differing outputs in v5 is not the same as Qwen3.5 save_pretrained writing incorrect visual encoder keys." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MOE conversion/loading failure is unrelated to Smollm3 RoPE layer dropping." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": false, + "reason": "General video regression across several models is too broad to merge with the specific qwen2.5-vl vision_position_ids issue." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "gguf_file breaks AutoTokenizer/AutoModel loading, while caching_allocator_warmup fails on Apple Silicon with TypeError; different code paths." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Physical-intelligence/fast loading failure is unrelated to CLIPTokenizer's pathological model_max_length." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Qwen3 auto device mapping cudaErrorAssert and Qwen3OmniMoe video input ValueError are different runtime failures." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "init_empty_weights _is_hf_initialized argument issue is unrelated to PIL backend processors requiring torchvision." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure and MobileLLM-125M key error are distinct loading/config bugs." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Loading physical-intelligence/fast on transformers>=5.1.0 is unrelated to list_repo_files not forwarding revision in CI." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "MacOS bus error on CLIP model load is unrelated to TrainerState token-count tracking." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:28282", + "issue:29127", + "issue:34567", + "issue:34689", + "issue:36331", + "issue:38175", + "issue:42175", + "issue:42915", + "issue:43317", + "issue:43452", + "issue:43493", + "issue:43526", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43643", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43782", + "issue:43824", + "issue:43906", + "issue:43927", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44737", + "issue:44821", + "issue:44898", + "issue:44912", + "issue:44938", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45137", + "issue:45325", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45405", + "issue:45479", + "issue:45538", + "issue:45588", + "issue:45823", + "issue:45920" + ], + "soft_pairs": [ + "issue:42915|issue:44821", + "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:43577|issue:45072", + "issue:44062|issue:44991", + "issue:42175|issue:44938", + "issue:38175|issue:44898", + "issue:42175|issue:44488", + "issue:45137|issue:45405", + "issue:43906|issue:45381", + "issue:44737|issue:45588", + "issue:28282|issue:34689", + "issue:42915|issue:43749", + "issue:43742|issue:44291", + "issue:43526|issue:45479", + "issue:43582|issue:43782", + "issue:43540|issue:45823", + "issue:43756|issue:45325", + "issue:28282|issue:29127", + "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" + ], + "prepared_review_unit_hash": "27922551fc9691a35b2ee5705e8d84f5adb6ab6923711f46e94d071bcf194b92", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 37039, + "estimated_input_tokens": 9260, + "estimated_eval_tokens": 18776 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are unrelated bug reports across different models/subsystems. I found one plausible duplicate pair around SigLIP2; the rest should stay separate.", + "confidence": 0.87, + "canonical_issue_reason": "No single canonical issue for the full set; the only believable duplicate pair is the SigLIP2 reports, where 43994 is the clearer anchor because it captures the user-visible failure and broader impact.", + "canonical_pr_reason": null, + "best_issue_reason": "43994 is the best overall representative among these issues because it describes a broad, user-facing regression in a widely used model path, while most others are narrower or unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Different models and failures: Qwen3Moe FP8 loading vs AutoImageProcessor URL loading." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated bugs: Qwen3Moe FP8 config error vs CLIPTokenizer max_length behavior." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "Different code paths: config save/load token IDs vs MXFP4 quantization fallback." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Device-map/offload loading bug is unrelated to CLIPTokenizer max_length." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Backend packaging issue vs missing Qwen2.5-VL class import." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related, but they target different Qwen model families and failure modes; not enough evidence of one underlying bug." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Different video-input errors in different models; not the same concrete bug." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype, but they affect different architectures and symptoms." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "Tokenizer errors, but one is AddedToken argument handling and the other is loading a specific tokenizer." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Backend installation mismatch is unrelated to Python 3.14 import/load failure." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Different models and failures: SigLIP2 zero probabilities vs Perceiver non-default resolution." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "Packaging/backend issue vs model loading failure for a specific checkpoint." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 runtime error is unrelated to PEFT version bump." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "No clear shared bug: isolated reproduction of an old issue vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet CPU placement bug is unrelated to flash_attention s_aux=None crash." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Generic PyTorch ImportError vs a specific Llama 3.2 Vision loading regression." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Different training/loading problems with no shared failing code path." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "Both are load-time issues, but one is a missing key in MobileLLM and the other is an unexpected init_empty_weights argument." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "Image processor label reduction bug vs sequence-classification zero-loss bug." + }, + { + "left": "issue:43582", + "right": "issue:43782", + "accept": false, + "reason": "Apple Silicon allocator TypeError is unrelated to Qwen3VL weight_only loading failure." + }, + { + "left": "issue:43540", + "right": "issue:45823", + "accept": false, + "reason": "Video processing error vs Gemma4 PLE device mismatch; different subsystems." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Different RoPE-related model bugs with different symptoms and affected models." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "PyTorch missing import error is unrelated to LayoutLMv3 box-validation messaging." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "Both are SigLIP2 regressions describing the same underlying model discrepancy, with 43994 showing the user-facing symptom and 43493 giving the implementation-vs-JAX diagnosis." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "TrainerState token counting bug vs CustomTrainer compute_loss signature mismatch." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer failures: AddedToken special-arg duplication vs Mistral regex patch crash." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "Model init-empty-weights argument bug is unrelated to AutoImageProcessor URL loading." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "save_pretrained visual-key regression is not the same as AutoTokenizer wrong token IDs." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both involve config fields, but one is trust_remote_code AutoConfig loss and the other is a specific Qwen vision config field." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "GGUF tokenizer/model loading breakage is unrelated to the sleng-bert loading report." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. It correctly treats most pairs as unrelated and only tentatively groups the SigLIP2 issues, which is the one pair whose titles strongly align. I don\u2019t see any overstatement that would require rejection." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:24643", + "issue:29127", + "issue:30064", + "issue:33666", + "issue:38175", + "issue:41762", + "issue:43295", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43643", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43819", + "issue:44361", + "issue:44442", + "issue:44464", + "issue:44466", + "issue:44484", + "issue:44514", + "issue:44561", + "issue:44568", + "issue:44589", + "issue:44625", + "issue:44704", + "issue:44743", + "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:45381", + "issue:45405", + "issue:45600", + "issue:45823" + ], + "soft_pairs": [ + "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:43819|issue:44811", + "issue:43295|issue:45356", + "issue:44936|issue:45823", + "issue:43653|issue:44568", + "issue:44464|issue:44743", + "issue:38175|issue:43295", + "issue:43643|issue:44561", + "issue:43653|issue:44361", + "issue:44361|issue:44442", + "issue:30064|issue:33666" + ], + "prepared_review_unit_hash": "ed851354ef47701a33e732a66e00c4f0b7509399f5561e92cac6613d8ff85ae2", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35983, + "estimated_input_tokens": 8996, + "estimated_eval_tokens": 18248 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are unrelated single-issue reports. The only plausible duplicate/near-duplicate pair is the tied-weights save/serialization regression in v5. No PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "issue:44466 is the broadest, most concrete report in the only likely duplicate family: it describes the tied-weights serialization problem in v5 in a way that can absorb model-specific variants.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44466 is the best representative issue for triage because it focuses on the underlying save_pretrained/tied-weights regression rather than a narrow model instance.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Different failures: a janus image-generation test case versus loading Phi-4 multimodal models. No shared code path." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Unrelated subsystems: PIL image processor torchvision dependency vs Mistral regex/tokenizer loading crash." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "Different model/load-time errors: MobileLLM key error versus Float8 storage lookup failure." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to loading AutoImageProcessor from a URL." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "One is a dependency version bump issue; the other is stale auto_mappings references to removed configs." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "Different bugs: RTDetrV2 output changes versus Qwen3.5 save_pretrained regression." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Both involve training/runtime infrastructure, but one is auxiliary-loss normalization and the other is FSDP loading." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Different chat-template/video-position-id problems; not the same concrete bug." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "BitNet accelerated-loading unpacking bug versus Perceiver image classification interpolation failure." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "save_pretrained shard size question is unrelated to AutoProcessor kwargs forwarding." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Gemma3 DeepSpeed ZeRO-3 load failure is not the same as trainer.evaluate() failing after train()." + }, + { + "left": "issue:24643", + "right": "issue:29127", + "accept": false, + "reason": "DeepSpeed weight-dimension runtime error is unrelated to LayoutLMv3 box validation messaging." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3.5 num_labels propagation is unrelated to granite_speech strict-config loading." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both are apply_chat_template-related, but they fail on different inputs and code paths." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Tokenizer/model loading via gguf_file is unrelated to Blip2 dtype propagation." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Different tokenizer-loading failures: gguf_file support versus offline model_info lookup in mistral regex patching." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": true, + "reason": "Both report the same tied-weights regression in v5 save/serialization behavior, with 45005 appearing model-specific and 44466 describing the underlying device-dependent bug." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC latent conversion mismatch and Whisper batch_decode special-token handling are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Both mention tokenizer/processor regressions, but one is custom processor.tokenizer access while the other is Kimi codec handling and warning behavior." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer lifecycle failure is unrelated to Gemma4 PLE device-map forward mismatch." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "BigBird special-token registration bug is unrelated to mdeberta-v3 BOS/EOS addition regression." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Different generation bugs: compiled forward chunking versus Qwen3.5 recurrent state reset with cache." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Siglip2 zero-probabilities issue is unrelated to processor.tokenizer regression." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both involve trust_remote_code, but one is missing AutoConfig fields and the other is v5 removal of is_torch_fx_available." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "Different tokenizer failures: BigBird mask-token registration versus MLukeTokenizer AttributeError." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "MLukeTokenizer task AttributeError is unrelated to FastSpeech2ConformerTokenizer loading." + }, + { + "left": "issue:30064", + "right": "issue:33666", + "accept": false, + "reason": "Void segmentation map processing and Qwen2-VL multi-GPU training are different problems." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the only accepted soft-edge pair is a conservative, plausible tied-weights duplicate family. The other verdicts stay appropriately cautious and do not overstate shared code paths or bug identity." + }, + "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": 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-10000-400", + "nodes": [ + "issue:28282", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:34567", + "issue:35141", + "issue:36010", + "issue:41720", + "issue:43122", + "issue:43278", + "issue:43452", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43825", + "issue:43906", + "issue:43927", + "issue:44361", + "issue:44451", + "issue:44464", + "issue:44479", + "issue:44554", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44821", + "issue:44843", + "issue:44849", + "issue:44977", + "issue:44991", + "issue:45357", + "issue:45440", + "issue:45588", + "issue:45701", + "issue:45704", + "issue:45812" + ], + "soft_pairs": [ + "issue:41720|issue:43906", + "issue:43688|issue:45440", + "issue:43688|issue:45704", + "issue:43577|issue:43697", + "issue:28282|issue:32090", + "issue:28282|issue:33357", + "issue:28282|issue:35141", + "issue:43646|issue:43749", + "issue:44843|issue:44991", + "issue:43688|issue:45701", + "issue:43653|issue:43927", + "issue:43531|issue:43742", + "issue:43550|issue:43927", + "issue:44464|issue:44977", + "issue:44479|issue:44625", + "issue:43550|issue:44361", + "issue:43577|issue:44554", + "issue:43577|issue:45588", + "issue:43122|issue:43906", + "issue:43504|issue:43550", + "issue:43504|issue:43720", + "issue:32090|issue:34567", + "issue:41720|issue:44464", + "issue:43504|issue:44610", + "issue:32090|issue:33666", + "issue:43452|issue:44451", + "issue:44849|issue:45357", + "issue:43742|issue:44821", + "issue:43278|issue:43688", + "issue:43493|issue:43825", + "issue:33666|issue:36010", + "issue:43606|issue:43927", + "issue:43720|issue:43927", + "issue:44568|issue:45812", + "issue:43525|issue:43572" + ], + "prepared_review_unit_hash": "2b5359add63ec35c01e6822a4d9bf03bbcedf01e415c3393aeb2b1ad78d477c4", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38168, + "estimated_input_tokens": 9542, + "estimated_eval_tokens": 19340 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous cluster spanning unrelated tokenizer, model-loading, training, generation, and config bugs. None of the proposed soft pairs look like the same underlying bug/change, so all soft edges are rejected.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:45812 is the strongest standalone representative: it is a concrete, reproducible tokenizer regression with broad impact across Granite models and clear v4\u2192v5 behavior change.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41720", + "right": "issue:43906", + "accept": false, + "reason": "Different failures: auto device mapping CUDA assert on A800 vs an isolated reproduction of another issue." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Different bugs and code paths: auxiliary-loss normalization vs native DeepSeekV3 divergence." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Different model/code paths: expert-loss normalization vs T5 apex FusedRMSNorm memory leak." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "Different symptoms and models: Blip2 dtype remaining fp32 vs RTDetrV2 output differences." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for missing PyTorch vs Trainer GPU broadcast NoneType TypeError are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Missing PyTorch ImportError vs MacOS CLIP bus error are unrelated failures." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "Different bug classes: dependency import failure vs embedding resize reinitialization." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Both are v5/loading-related, but they target different mechanisms: custom init breakage vs FSDP CPU RAM efficient loading." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Different tokenizer paths: unconditional model_info call in Mistral regex patch vs tokenizer loading for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Auxiliary-loss normalization and tokenization changes are unrelated problems." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "BigBirdTokenizer special-token registration vs DiaConfig save/load token-ID loss are different bugs." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3-MoE sliding_window issue and MobileLLM load key error are different code paths." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "torch.compile/SDPA behavior in Bamba is unrelated to DiaConfig token-ID persistence." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Chunked generation with compiled forward vs Qwen3.5 flash-attention generation failure are different issues." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression for Qwen VL models vs num_labels propagation in Qwen3.5 are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "Bamba compile/SDPA bug vs MLukeTokenizer AttributeError are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "Blip2 dtype handling and an MPS attention correctness bug are different problems." + }, + { + "left": "issue:43577", + "right": "issue:45588", + "accept": false, + "reason": "Blip2 dtype retention vs flash_attention.py crashing on s_aux=None are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:43906", + "accept": false, + "reason": "Different tokenization regression reports with no clear shared code path or concrete duplicate evidence." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Beit legacy-field loading failure vs Bamba torch.compile/SDPA bug are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Beit preset loading and BitNet packed-weight unpacking are different load-time bugs." + }, + { + "left": "issue:32090", + "right": "issue:34567", + "accept": false, + "reason": "Trainer broadcast NoneType error vs num_input_tokens_seen not updating are separate Trainer issues." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "CUDA assert during auto device mapping is unrelated to chunked generation inconsistency." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Legacy config field loading vs processor/model input-size mismatch are different bugs." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer broadcast NoneType error and Qwen2-VL multi-GPU training are not the same failure." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "GGUF/from_pretrained breakage and a specific model-load failure for ScandiBERT are different issues." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3.5 hidden-states bug vs save_pretrained visual-encoder key regression are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "MobileLLM key error vs AutoImageProcessor URL loading failure are different loading paths." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype drift in evaluate vs auxiliary-loss normalization are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy vs pipeline translation error-message regression are different concerns." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and GenerationMixin import failure do not share the same bug." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU-offload device mismatch for bark-small vs DiaConfig token-ID loss are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "BitNet accelerate-loading unpacking bug vs DiaConfig token-ID persistence bug are different." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and code paths: special-token injection for mdeberta-v3 vs wrong token IDs for Granite." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config and missing pad_token_idx in StableLmConfig are similar in shape but not the same concrete bug or fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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-10000-400", + "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:43540", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43653", + "issue:43720", + "issue:43756", + "issue:43824", + "issue:43844", + "issue:43927", + "issue:43994", + "issue:44112", + "issue:44291", + "issue:44361", + "issue:44423", + "issue:44464", + "issue:44466", + "issue:44488", + "issue:44493", + "issue:44589", + "issue:44625", + "issue:44734", + "issue:44843", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45405", + "issue:45520", + "issue:45584", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "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:43278|issue:43844", + "issue:43334|issue:43572", + "issue:44625|issue:45584", + "issue:43257|issue:43299", + "issue:43756|issue:45083", + "issue:43334|issue:45759", + "issue:44488|issue:44843", + "issue:45200|issue:45229", + "issue:43824|issue:44589", + "issue:43646|issue:44466", + "issue:39692|issue:43295", + "issue:45229|issue:45823", + "issue:43504|issue:43606", + "issue:44589|issue:45520", + "issue:43572|issue:44291", + "issue:33357|issue:33666", + "issue:44423|issue:44734", + "issue:43927|issue:44361", + "issue:43504|issue:43653", + "issue:44464|issue:45584", + "issue:43606|issue:44112", + "issue:43994|issue:44493", + "issue:43575|issue:45405" + ], + "prepared_review_unit_hash": "9fff9d47fd7ccaa3e4cfbc8ef9993c69c2f175e6b0b20c345c2c816c5b112857", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37543, + "estimated_input_tokens": 9386, + "estimated_eval_tokens": 19028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous and the soft pairs are only broad thematic matches, not true duplicates. I would reject all proposed duplicate edges and avoid selecting a canonical issue/PR for the cluster.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio sample-rate default bug vs processor/tokenizer regression; different subsystems and failure modes." + }, + { + "left": "issue:43756", + "right": "issue:44913", + "accept": false, + "reason": "Both mention rotary settings, but one is a Smollm3 RoPE-layer implementation issue and the other is a GPTNeoX config reload bug." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet packed-weight loading failure vs MLukeTokenizer task-time AttributeError; unrelated code paths." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer spacing bug vs MacOS bus error on a CLIP model; not the same defect." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Qwen2-VL multi-GPU training request vs custom trainer loss-signature TypeError; different problems." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Documentation/example errors vs SigLIP2 implementation parity discrepancy; not the same underlying bug." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3OmniMoe video-input processing vs Qwen3.5 flash-attention generation issue; different code paths." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 model mismatch issue vs DeepSpeed/gradient behavior regression; unrelated." + }, + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Phi-4 multimodal loading failure vs PEFT version pin bump; no shared bug." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Eval/train dtype mismatch vs ZeRO-3 gradient inflation; different defects." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both involve missing pad-token fields, but in different models/config layers and not the same concrete bug." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Qwen3.5 num_labels propagation bug vs Whisper empty-transcription generation failure; unrelated." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both are Qwen MOE-related, but one is accelerate/deepspeed weight conversion and the other is multimodal loading; not mergeable as one fix." + }, + { + "left": "issue:43756", + "right": "issue:45083", + "accept": false, + "reason": "Smollm3 RoPE-layer mismatch vs qwen3_omni_moe feature-length helper behavior; different bugs." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_id attribute vs unwrap text_config in from_config; related family, but distinct code paths." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Model load failure for a specific model vs offline-mode model_info call; unrelated." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Text-only token-type defaulting vs multi-GPU inference OOM; different issues." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "ImportError for a Qwen2.5-VL class vs missing Float8 storage type; unrelated failures." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization regression vs device-dependent serialization inconsistency; not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "SigLIP2 docs/example problems vs processor/tokenizer regression; different scopes and code paths." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Both Gemma4-related, but one is CUDA OOM and the other is a device mismatch during forward; not the same defect." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Beit legacy-field loading bug vs Bark CPU-offload device mismatch; unrelated." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure vs flash_attn import-utils KeyError on Python 3.13; different issues." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Both are v5 init/config regressions, but one is a StableLmConfig field issue and the other is an init_empty_weights argument change." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS CLIP bus error vs Qwen2-VL multi-GPU training; no common bug." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Serve crash on multimodal model input conversion vs KV-cache continuation tensor indexing; different serve paths." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig save/load token-ID loss vs MLukeTokenizer AttributeError; unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Legacy-field model load bug vs BigBirdTokenizer special-token registration issue; different fixes." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Chunked generation inconsistency under compiled forward vs Whisper empty transcription after align_special_tokens; unrelated." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "Real CPU-offload device mismatch bug vs a stale GraniteSpeech CI test; not duplicates." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 pipeline output issue vs broad unexpected position-id keys; not the same underlying problem." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Tensor-parallel OOM on a specific model vs unreleased PEFT version bump; 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": 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-10000-400", + "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:43540", + "issue:43572", + "issue:43582", + "issue:43696", + "issue:43701", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43957", + "issue:44315", + "issue:44464", + "issue:44466", + "issue:44521", + "issue:44554", + "issue:44610", + "issue:44792", + "issue:44849", + "issue:44898", + "issue:44936", + "issue:44964", + "issue:44977", + "issue:44987", + "issue:45072", + "issue:45084", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45325", + "issue:45356", + "issue:45405", + "issue:45412", + "issue:45446", + "issue:45520", + "issue:45563", + "issue:45823" + ], + "soft_pairs": [ + "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:44554|issue:45356", + "issue:43493|issue:43701", + "issue:43295|issue:44610", + "issue:43873|issue:44464", + "issue:44898|issue:45412", + "issue:44466|issue:45127", + "issue:44792|issue:45563", + "issue:43572|issue:44987", + "issue:45200|issue:45325", + "issue:37428|issue:43824", + "issue:43572|issue:43957", + "issue:43844|issue:44936" + ], + "prepared_review_unit_hash": "eee1573adc68cd615b64d0c7f0f40d98a8056232111f73de13044540dfe69725", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36425, + "estimated_input_tokens": 9107, + "estimated_eval_tokens": 18470 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft links mostly share broad areas like tokenization, generation, or multimodal models, but they do not describe the same concrete bug or change. No duplicate canonical issue/PR stands out.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "BF16->FP32 dtype drift in evaluation and resume_from_checkpoint key mismatch are different failures with different code paths." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "A Janus model test failure and an unreleased PEFT version bump are unrelated." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both involve chat templating, but one is wrong assistant masks for multimodal inputs and the other is a crash on tool-call assistant messages; different bugs." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention 2 test failures and MLflow reporting after job failure are unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Local-path import warnings and CUDA OOM loading GPT-oss-20b are not the same issue." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient growth and a model load/import failure are different problems." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing after added tokens and TrainerState token counters not updating are separate behaviors." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "An MPS attention correctness bug and a PyTorch version check for AuxRequest import are unrelated." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Template-node compilation errors and a flash_attn import KeyError on Python 3.13 are distinct import/compile paths." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon warmup TypeError and Qwen3.5 flash-attention generation issues are not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Video-processing ValueError in Qwen3OmniMoe and Gemma4 text-only mm_token_type_ids defaults are different multimodal issues." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero probabilities are unrelated model bugs." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation kwargs after sync_gpus and a Qwen3.5 hidden-states bug do not share the same underlying defect." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "bfloat16 dtype mismatches in inference and wav2vec2 save/tokenization failures are different code paths." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "Missing flash_supports_window_size import and meta-device loading regressions are separate import/model-loading bugs." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Gemma4 mm_token_type_ids defaults and device-map forward mismatches are not the same problem." + }, + { + "left": "issue:44315", + "right": "issue:44849", + "accept": false, + "reason": "Liger Kernel application with model_init and Qwen3.5 hidden-state behavior are unrelated." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "MPS attention correctness and Kimi-K2.5 tokenizer regression are different subsystems and failures." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 HF-vs-JAX discrepancy and checkpoint resume key mismatch are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Processor/tokenizer regression and OmDet-Turbo input-size mismatch are different bugs." + }, + { + "left": "issue:43873", + "right": "issue:44464", + "accept": false, + "reason": "Quantization offloading behavior and chunked generation under compiled forward are not the same issue." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver resolution interpolation failure and RT-DETR memory release leak are unrelated." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "lm_head.weight serialization inconsistency and LoRA merge collapse with extended vocab are different failure modes." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "A Janus model test failure and a stale warning in paged generate() are not the same bug." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "Missing pad_token_idx in config and a transformers>=5.1.0 load failure are only loosely related versioned-model issues, not one concrete defect." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Gemma4 token-type defaults and Qwen2.5-VL rope index scaling are different multimodal bugs." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "These are both import errors, but they concern different missing symbols for different models." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "A missing config field and a meta-device loading regression are not the same underlying change." + }, + { + "left": "issue:43844", + "right": "issue:44936", + "accept": false, + "reason": "ZeRO-3 gradient instability and trainer.evaluate() failing after train() are different training problems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43335", + "issue:43388", + "issue:43425", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43825", + "issue:43827", + "issue:43856", + "issue:43873", + "issue:43901", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44442", + "issue:44448", + "issue:44514", + "issue:44521", + "issue:44610", + "issue:44743", + "issue:44857", + "issue:45083", + "issue:45127", + "issue:45198", + "issue:45245", + "issue:45290", + "issue:45381", + "issue:45412", + "issue:45584", + "issue:45593" + ], + "soft_pairs": [ + "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:44514|issue:44521", + "issue:38175|issue:43493", + "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", + "issue:43278|issue:43825", + "issue:37428|issue:43425", + "issue:43606|issue:45083", + "issue:45290|issue:45381", + "issue:43335|issue:43550", + "issue:44448|issue:44743", + "issue:44857|issue:45593" + ], + "prepared_review_unit_hash": "dd08ec80d0a6a3130b7b16dd0883e6c98f47312e92c9838e41a27d80e6178b7d", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35995, + "estimated_input_tokens": 8999, + "estimated_eval_tokens": 18254 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly heterogeneous: tokenization, multimodal chat-template, import, runtime, and training issues that only look superficially similar. I found one plausible duplicate family around missing `pad_token_id` on config objects; the rest should stay separate.", + "confidence": 0.81, + "canonical_issue_reason": "issue:43334 is the clearest anchor for the only plausible duplicate family here: an `AttributeError` from missing `pad_token_id` on config objects, with a near-identical sibling in issue:43525.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43334 is the best overall issue to anchor the cluster because it has a concrete, model-loading code-path bug and a plausible duplicate; the other items are mostly unrelated or only broadly similar.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Different regressions: processor/tokenizer API breakage vs batched video shape handling." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": true, + "reason": "Both report the same underlying `pad_token_id` config attribute bug causing load-time `AttributeError`." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "Unrelated failures: an import error for a missing class vs a detokenization/output issue." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Processor regression and quantization/offloading behavior are different code paths and symptoms." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Different bugs in different helpers: metric gathering label truncation vs image processor label reduction." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 output anomaly and RT-DETR memory retention are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "FP8 config failure and Qwen2.5-VL import failure are different issues." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both involve `apply_chat_template`, but one crashes on batched input and the other produces incorrect assistant masks; not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "Same model family, but the titles point to different symptoms and not clearly the same fix." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "Tokenizer load failure vs save/tokenization failure in a different model family." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing behavior and embedding reinitialization are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Training/evaluation dtype change and MoE memory usage are distinct problems." + }, + { + "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 vs a stale CI device-override test." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs-related, but they cover different API deprecations and different pages." + }, + { + "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": "Feature-extractor center-arg crash and Whisper empty-transcription handling are different failures." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/vocabulary merge collapse and category-count overflow are not the same bug." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC STE mismatch and qwen3_omni_moe length helper behavior are different code paths." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC forward mismatch and a GraniteSpeech CI test issue are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sample-rate defaulting and video batching shape are different subsystems." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor input-size mismatch and AMP/CUDA loss crash are separate bugs." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype regression and pipeline docs mismatch are unrelated." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "A specific import error and a Torch version incompatibility are not the same issue." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU offload device mismatch and qwen3_omni_moe feature-length helper behavior are unrelated." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat-template tool-call crash and video vision-position-id bug are different multimodal issues." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "Sparse-layer construction bug and Bamba SDPA/compile failure do not share the same root cause." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Output mismatch for Pegasus and recurrent-state reset in modular Qwen3 are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "AMP/CUDA crash and missing auxiliary losses are different model-training bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analysis is grounded in the packet and stays conservative. The anchor choice around issue:43334 is supported by the matching `pad_token_id` AttributeError in issue:43525, and the summary does not overclaim broader duplication beyond that plausible family. The other pairwise rejections are phrased cautiously and are consistent with the issue titles." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:32090", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:39692", + "issue:40444", + "issue:42490", + "issue:43335", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43720", + "issue:43723", + "issue:43819", + "issue:43854", + "issue:43931", + "issue:44112", + "issue:44206", + "issue:44292", + "issue:44351", + "issue:44360", + "issue:44448", + "issue:44462", + "issue:44493", + "issue:44554", + "issue:44561", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45310", + "issue:45478", + "issue:45520", + "issue:45563", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:44360|issue:44561", + "issue:45200|issue:45588", + "issue:37428|issue:44351", + "issue:29942|issue:32090", + "issue:44206|issue:45823", + "issue:43611|issue:43931", + "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:38175|issue:44493", + "issue:43335|issue:45072", + "issue:44448|issue:44779", + "issue:44462|issue:44561", + "issue:44561|issue:44779", + "issue:43335|issue:45593", + "issue:43720|issue:44112", + "issue:44977|issue:45229", + "issue:42490|issue:43854", + "issue:40444|issue:43526", + "issue:44448|issue:44977", + "issue:34689|issue:35141", + "issue:43606|issue:44857", + "issue:43723|issue:44554", + "issue:44857|issue:45072", + "issue:44977|issue:45563" + ], + "prepared_review_unit_hash": "2f25421e8c38744aae2da9fb5ff9b1f023f6ad5de640618830760843e9e5fc3c", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "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": "Only one soft edge looks like a true duplicate pair; the rest share keywords but describe different bugs, regressions, or subsystems.", + "confidence": 0.97, + "canonical_issue_reason": "issue:45310 is the earlier, fuller report of the same Qwen3.5 MoE `from_pretrained` regression as issue:45478.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the clearest duplicate target in this set: specific, reproducible, and matched by the near-identical follow-up report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "Different code paths: DSA indexer/ReLU discussion vs v5 trust_remote_code breakage from `is_torch_fx_available` removal." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Different failures: missing `mm_token_type_ids` for Gemma 4 fine-tuning vs flash-attention crash on `s_aux=None`." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different missing symbols (`_flash_supports_window_size` vs `HybridCache`)." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures are unrelated to the Trainer `_gpu_broadcast_one` NoneType broadcast error." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Different subsystems and symptoms: feature extractor `center` arg crash vs Gemma4 device-map mismatch during forward." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Model-loading regression from `base_model_prefix` handling is distinct from Qwen3-VL weight-shape mismatch." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "SigLIP2 docs/example issues are unrelated to BeitImageProcessorFast `reduce_labels` returning one label." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug is unrelated to BigBirdTokenizer special-token decode behavior." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC `from_latents`/STE mismatch is a different bug from unexpected `position_id` keys across models." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Different targets: Bamba torch.compile+SDPA failure vs stale GraniteSpeech CI test/device override issue." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "Same broad model family, but different bugs: sparse-layer creation vs Beit legacy-field loading preset failure." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 JIT parse/indentation import error is unrelated to qwen3_omni_moe feature-length helper behavior." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Near-identical reports of the same Qwen3.5 MoE `from_pretrained` regression on transformers>=5.4.0." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "SwitchTransformers config bug and qwen3_omni_moe feature-length helper bug are unrelated." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMV3 box/clarity error is unrelated to Trainer `_gpu_broadcast_one` NoneType broadcast failure." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Different flash-attention edge cases: sink-less `s_aux=None` crash vs CPU-offload incompatibility in Qwen3_5MoE rotary embedding." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 documentation/model mismatch is unrelated to unexpected `position_id` keys." + }, + { + "left": "issue:44292", + "right": "issue:45520", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime failure is unrelated to the Python 3.13 `flash_attn` import key error." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Zero-probability outputs in SigLIP2 are a different symptom from unexpected `position_id` key warnings." + }, + { + "left": "issue:43335", + "right": "issue:45072", + "accept": false, + "reason": "Different failure modes: SwitchTransformers sparse-layer config vs dtype mismatches in bfloat16 inference." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Different regressions in different models: Pegasus v4/v5 output mismatch vs Deepseek tokenizer correctness change." + }, + { + "left": "issue:44462", + "right": "issue:44561", + "accept": false, + "reason": "AutoTokenizer ignoring `tokenizer.json` is unrelated to trust_remote_code breakage from removed `is_torch_fx_available`." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Different areas: remote-code utility removal vs Deepseek tokenizer output regression." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug is unrelated to D-FINE auxiliary-loss handling." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "BitNet accelerate-loading unpack bug is unrelated to GraniteSpeech stale CI device-override test." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug is unrelated to Gemma4 multi-GPU CUDA OOM." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Trailing-slash local model import warning is unrelated to GLM-4.7-Flash loading failure in tests." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "IterableDataset multi-image finetuning failure is unrelated to BeitImageProcessorFast `reduce_labels` bug." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Different model/output regressions: Pegasus v4/v5 mismatch vs Qwen3.5 flash-attention generation issue." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Llama 3.2 Vision loading breakage and token-embedding reinit on `post_init` are different bugs." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU offload device mismatch in bark-small is unrelated to LwDetrImageLoss AMP/CUDA crash." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "Tokenizer loading in v5 and MPS attention correctness are different problems." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Different failures: loss/AMP crash vs dtype mismatch in inference." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Qwen3.5 flash-attention generation bug is unrelated to the stale `num_return_sequences` warning in paged generate()." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The canonical duplicate choice is well grounded: issue:45310 and issue:45478 have effectively the same title and clearly refer to the same Qwen3.5 MoE from_pretrained regression. The remaining soft-edge rejections are conservative and supported by the packet-level titles/symptoms, without overstating 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": 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-10000-400", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:28218", + "issue:29127", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:35141", + "issue:37428", + "issue:41720", + "issue:42490", + "issue:43065", + "issue:43278", + "issue:43329", + "issue:43381", + "issue:43412", + "issue:43550", + "issue:43575", + "issue:43653", + "issue:43701", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:43994", + "issue:44448", + "issue:44466", + "issue:44493", + "issue:44568", + "issue:44743", + "issue:44857", + "issue:44987", + "issue:45083", + "issue:45198", + "issue:45229", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45712", + "issue:45803", + "issue:45823" + ], + "soft_pairs": [ + "issue:44568|issue:45491", + "issue:43927|issue:45698", + "issue:28218|issue:32090", + "issue:43701|issue:45083", + "issue:43550|issue:45083", + "issue:45198|issue:45803", + "issue:45372|issue:45588", + "issue:37428|issue:44987", + "issue:29942|issue:35141", + "issue:43873|issue:44493", + "issue:45459|issue:45712", + "issue:32090|issue:33357", + "issue:45245|issue:45563", + "issue:43412|issue:45588", + "issue:43873|issue:45229", + "issue:41720|issue:45229", + "issue:43278|issue:43329", + "issue:43976|issue:45440", + "issue:10000|issue:15354", + "issue:45372|issue:45406", + "issue:43065|issue:43994", + "issue:42490|issue:43575", + "issue:28218|issue:29127", + "issue:43065|issue:45412", + "issue:45372|issue:45459", + "issue:44857|issue:45588", + "issue:43653|issue:44743", + "issue:44448|issue:44466", + "issue:43278|issue:43381", + "issue:45588|issue:45823" + ], + "prepared_review_unit_hash": "8e9fcfb3eaf43d6f226da521c48ee5d0c3fc8045d73de8800f7e9f5a569d8b78", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36365, + "estimated_input_tokens": 9092, + "estimated_eval_tokens": 18440 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is broadly heterogeneous. A few items share themes like tokenizer/save-load regressions or Gemma4 processor-loading failures, but none of the soft pairs look like the same concrete bug/change, so I reject all duplicate candidates.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45372 is the most central representative: it is a current Gemma4 processor-loading regression, has downstream references, and sits closest to the small number of related Gemma4/processor items in the set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45372 is the best single anchor for this set because it is specific, recent, open, and most likely to attract related follow-up reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer special-token behavior vs Gemma3 NaN embeddings are unrelated bugs." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save/load flows, but one is config token IDs and the other is wrong custom module resolution." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing after added tokens is unrelated to Trainer broadcast NoneType errors." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint key mismatch and multimodal feature-length behavior are different code paths." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile/SDPA failure is unrelated to the feature-length helper issue." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure and Fuyu bare-except masking are distinct tokenizer issues." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Both touch Gemma4-related loading/runtime, but one is an import dependency break and the other is a flash-attention attribute crash." + }, + { + "left": "issue:37428", + "right": "issue:44987", + "accept": false, + "reason": "Missing flash-attention utility import is not the same as failing to load a specific model repository." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Flash Attention 2 test failures are unrelated to embedding reinitialization on resize." + }, + { + "left": "issue:43873", + "right": "issue:44493", + "accept": false, + "reason": "Quantized offloading behavior and unexpected position-id keys are different defects." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Masked tokenizer decode errors and leftover dummy classes are unrelated maintenance issues." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer NoneType broadcast error is unrelated to a MacOS bus error on a CLIP model." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Category-cardinality runtime error is unrelated to a stale warning in paged generate()." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export failures and flash-attention sink-less crash are different concrete problems." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Quantization/offloading behavior is not the same as multi-GPU OOM on Gemma4." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "A800 cudaErrorAssert with auto device mapping is not the same as a multi-GPU OOM." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype drift and undefined video-branch helpers are unrelated model issues." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility and DeepSeek implementation divergence are different failures." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "German DistilBertModel issue and torch.jit GeneratorExp export failure are unrelated." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Both involve Gemma4Processor, but the reported failures point to different root causes and code paths." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to SigLIP2 producing nonsensical outputs." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Trailing-slash import warning and tensor-parallel OOM are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:29127", + "accept": false, + "reason": "Tokenizer added-token spacing and LayoutLMv3 box-validation messaging are different issues." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "Sam3PixelDecoder dummy layer issue is unrelated to RT-DETR memory not releasing." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma4 processor import failure and protobuf-error masking in tokenizers are not the same bug." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "AMP/CUDA loss crash and flash-attention sinkless-model crash are different runtime failures." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "BigBird special-token registration and Qwen3 recurrent-state reset are unrelated." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Version output differences and lm_head serialization inconsistency are distinct model-state bugs." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Training/eval dtype change and gradient-checkpointing-in-eval are unrelated." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Flash-attention s_aux=None crash and Gemma4 PLE device mismatch are separate Gemma4 runtime 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-10000-400", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:36010", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43066", + "issue:43257", + "issue:43278", + "issue:43381", + "issue:43421", + "issue:43526", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44279", + "issue:44361", + "issue:44442", + "issue:44451", + "issue:44462", + "issue:44479", + "issue:44610", + "issue:44871", + "issue:44977", + "issue:45103", + "issue:45198", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45491", + "issue:45520", + "issue:45593", + "issue:45834", + "issue:45859" + ], + "soft_pairs": [ + "issue:44206|issue:44479", + "issue:43066|issue:44462", + "issue:43994|issue:45412", + "issue:25251|issue:32090", + "issue:44186|issue:45198", + "issue:44451|issue:45520", + "issue:42915|issue:42994", + "issue:44361|issue:45198", + "issue:44871|issue:44977", + "issue:44265|issue:45198", + "issue:43756|issue:45440", + "issue:43421|issue:45491", + "issue:29942|issue:30064", + "issue:43257|issue:43278", + "issue:33357|issue:39692", + "issue:44610|issue:45412", + "issue:29127|issue:34567", + "issue:42490|issue:43873", + "issue:32090|issue:33290", + "issue:45103|issue:45372", + "issue:45440|issue:45593", + "issue:43824|issue:44279", + "issue:43653|issue:44442", + "issue:43761|issue:45859", + "issue:43526|issue:45834", + "issue:43381|issue:43701", + "issue:43526|issue:44186", + "issue:29942|issue:36010", + "issue:29127|issue:33357" + ], + "prepared_review_unit_hash": "ee0a4c73a5d587a76ac5f59388e886383b5527958b8abbd48ef3f349c652a9d6", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36306, + "estimated_input_tokens": 9077, + "estimated_eval_tokens": 18410 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a broad set of mostly unrelated Transformers issues that only look superficially similar by text embedding. None of the proposed soft edges are true duplicates, so all should be rejected. There is no meaningful canonical issue for the cluster.", + "confidence": 0.92, + "canonical_issue_reason": "No canonical issue: the items span different subsystems and failure modes (tokenization, quantization, training, vision, docs, export, offload), so there is no single underlying bug to represent the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue: the soft matches are all broad similarity false positives rather than one coherent bug family, so no issue is a strong dedupe anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Different regressions in different model paths: LasrFeatureExtractor center-arg handling vs Qwen video-input processing." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Both involve tokenizers, but one is a decoder-type mismatch in v5 while the other is AutoTokenizer ignoring tokenizer.json; not the same bug." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "Unrelated failures: SigLIP2 inference corruption vs RT-DETR model memory not being released." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Top-k pipeline output nesting and trainer _gpu_broadcast_one NoneType errors are unrelated code paths." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "Both are tokenizer-related, but the concrete failures differ: LayoutLMv2 NER/padding crash vs Wav2Vec2 save_pretrained/tokenization failure." + }, + { + "left": "issue:44451", + "right": "issue:45520", + "accept": false, + "reason": "Different root causes: model loading failure for ScandiBERT vs Python 3.13 flash_attn import-key issue." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Both touch quantization, but one is Qwen3MoE with FineGrainedFP8Config and the other is quantized model saving; not the same defect." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "Different tokenizer/model failures affecting different tasks and APIs; no shared underlying bug is evident." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Eos-token config mismatch is not the same issue as Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to Wav2Vec2 tokenization/save_pretrained problems." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Both concern model implementation divergence, but they affect different architectures and different mechanisms; not mergeable as one fix." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Runtime special-token post-processing is unrelated to Gemma3 NaN embeddings from sliding-window attention." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures and void segmentation map image-processing failures are separate bugs." + }, + { + "left": "issue:43257", + "right": "issue:43278", + "accept": false, + "reason": "Different issues: MoE weight conversion with accelerate+deepspeed vs embedding dtype changing between train/eval." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "Both mention CLIP/SigLIP-family examples, but one is a Mac bus error on a community model and the other is a documentation example mismatch/quantization issue." + }, + { + "left": "issue:44610", + "right": "issue:45412", + "accept": false, + "reason": "Processor output size mismatch and RT-DETR memory leak are unrelated." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and TrainerState token-count tracking are different issues." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Local-path import warning and quantization offloading behavior are not the same bug." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Trainer NoneType broadcast error and DeepSpeed Adafactor OOM are different failure modes in different layers." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring annotation handling and Gemma 4 processor loading/import error are unrelated." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepseekV3MoE implementation drift is not the same as D-FINE auxiliary-loss logic when denoising is off." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "Qwen2.5-VL import failure is not the same as a generic dependency issue in transformers." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "BigBird mask-token registration/empty decode and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are different tokenizer bugs." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are unrelated model-forward issues." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels behavior and Kosmos2.5 OCR index errors are different image-processing problems." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Gradient checkpointing in eval mode and resume_from_checkpoint key mismatch are not the same defect." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "ImageProcessorFast label reduction and LayoutLMv2 tokenizer crashes are different components and failure modes." + }, + { + "left": "issue:29942", + "right": "issue:36010", + "accept": false, + "reason": "Flash Attention 2 test failures are unrelated to an ImportError for GenerationMixin." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and a MacOS CLIP bus error 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": 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-10000-400", + "nodes": [ + "issue:29127", + "issue:33290", + "issue:34567", + "issue:35141", + "issue:42490", + "issue:42907", + "issue:42915", + "issue:43232", + "issue:43262", + "issue:43316", + "issue:43334", + "issue:43381", + "issue:43388", + "issue:43441", + "issue:43525", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43701", + "issue:43723", + "issue:43761", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44188", + "issue:44206", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44479", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44898", + "issue:44977", + "issue:45003", + "issue:45072", + "issue:45245", + "issue:45290", + "issue:45440", + "issue:45479", + "issue:45491", + "issue:45588", + "issue:45698", + "issue:45803", + "issue:45907" + ], + "soft_pairs": [ + "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:43262|issue:44206", + "issue:43526|issue:44857", + "issue:42490|issue:43381", + "issue:44062|issue:45907", + "issue:43232|issue:43388", + "issue:43441|issue:45588", + "issue:44743|issue:44811", + "issue:43761|issue:44079", + "issue:43646|issue:44077", + "issue:43575|issue:43701", + "issue:45003|issue:45440", + "issue:43575|issue:44387", + "issue:43723|issue:45479", + "issue:44442|issue:45803", + "issue:43761|issue:44188", + "issue:44479|issue:45290", + "issue:43316|issue:43525", + "issue:42907|issue:45698", + "issue:44442|issue:44898", + "issue:42490|issue:44743", + "issue:33290|issue:34567" + ], + "prepared_review_unit_hash": "4dfbf2e175a9b36c154645749ed5b3373096da7a26b298968905913070f7320e", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "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": "The items are mostly unrelated one-off bugs/regressions across different subsystems; the soft pairs share superficial vocabulary but not the same concrete failure mode. I would reject all soft edges.", + "confidence": 0.96, + "canonical_issue_reason": "issue:45698 is the broadest, most general loading/saving bug in the set and is still open; the rest are mostly model-specific or subsystem-specific regressions that do not form a clear duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45698 is the best representative if one issue must anchor this cluster: it concerns core from_pretrained/save_pretrained behavior and has broader reuse potential than the other narrowly scoped bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Different Qwen issues: missing `pad_token_id` on Qwen3-VL vs `sliding_window` behavior on Qwen3-MoE; not the same bug." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention generation regression vs a generic category-count runtime limit; unrelated failure modes." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "LayoutLMv3 box-clarity error message vs token-embedding reinitialization after resize; different code paths and symptoms." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Compiled chunked generation inconsistency vs Gemma3 NaN embeddings from sliding-window/all-padding windows; not the same defect." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3Moe FineGrainedFP8Config failure vs Qwen3-VL weight-shape mismatch; separate model-loading problems." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure for FastSpeech2ConformerTokenizer vs dtype mismatch in SwitchTransformers/TimmWrapperModel inference." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate default issue vs unsupported `center` argument in a different feature extractor." + }, + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "BEiTImageProcessorFast label reduction bug vs LwDetrImageLoss AMP/CUDA crash; unrelated subsystems." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Trailing-slash local model import warning vs gradient checkpointing being disallowed in eval mode; not the same bug." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "`AddedToken` keyword collision vs `list_repo_files` not forwarding `revision`; different APIs and failures." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs update after `sync_gpus` vs metric gathering dropping tuple labels; separate bugs." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both involve flash attention, but one is a Ministral-3 model regression and the other is an `s_aux=None` crash in a different integration path." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Recurrent state reset in Qwen3.5 MoE vs Whisper `batch_decode` ignoring `skip_special_tokens`; no shared underlying bug." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel not returning hidden states vs ModelOutput key assignment when a key was previously `None`; different abstractions." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Transformers 5.0 custom model init breakage vs `patchtsmixer` `post_init` contract issue; not the same problem." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "TP loading OOM for a Qwen2 model vs checkpoint resume key mismatch; unrelated." + }, + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "Unsafe `sys.modules` access in `modeling_utils` vs DeepSeekV3MoE implementation divergence; different code paths." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both mention OOM, but one is TP model loading and the other is int4 quantization reserved-memory growth; not mergeable as one bug." + }, + { + "left": "issue:43723", + "right": "issue:45479", + "accept": false, + "reason": "AutoTokenizer loading issue in v5 vs degenerate zero-loss sequence classification; unrelated." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer loading failure vs bare `except` in Fuyu tensor conversion; no shared underlying defect." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "Hidden-states regression in CLIPVisionModel vs attention-kernel divergence under `torch.compile`; distinct issues." + }, + { + "left": "issue:44479", + "right": "issue:45290", + "accept": false, + "reason": "Video-input regression for Qwen-family models vs chat-template tokenization crash on tool-call assistant messages; different layers." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3TextConfig API discrepancy vs Llama4Config missing `pad_token_id`; similar theme, but not the same bug." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Both touch model save/load, but one is dequantized Ministral save failure and the other is wrong custom module loading after save/load; too different to merge." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading failure vs Perceiver non-default-resolution image classification failure; unrelated." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local path import warning vs Qwen3.5 recurrent-state reset bug; no common failure mode." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "DeepSpeed Adafactor OOM vs `num_input_tokens_seen` not updating; entirely separate training issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:25251", + "issue:28282", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:36010", + "issue:36331", + "issue:39692", + "issue:41628", + "issue:42175", + "issue:43065", + "issue:43295", + "issue:43316", + "issue:43381", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43493", + "issue:43540", + "issue:43611", + "issue:43643", + "issue:43673", + "issue:43688", + "issue:43720", + "issue:43749", + "issue:43825", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44077", + "issue:44162", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44617", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44871", + "issue:44877", + "issue:44913", + "issue:45137", + "issue:45216", + "issue:45372", + "issue:45491", + "issue:45759" + ], + "soft_pairs": [ + "issue:34689|issue:43611", + "issue:39692|issue:45491", + "issue:33290|issue:35141", + "issue:43931|issue:45216", + "issue:33290|issue:34689", + "issue:43381|issue:43873", + "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:44190|issue:44496", + "issue:33290|issue:36331", + "issue:44855|issue:45137", + "issue:43316|issue:44877", + "issue:43540|issue:44514", + "issue:42175|issue:43316", + "issue:43673|issue:44743", + "issue:43065|issue:44617", + "issue:43404|issue:43441", + "issue:44162|issue:44792", + "issue:43295|issue:43825", + "issue:25251|issue:28282", + "issue:43388|issue:43688" + ], + "prepared_review_unit_hash": "1a7b2f9b063d6aaf2222e6d3d8f35b2cb02f0997eaa0d4afcecb595a62dce6fa", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36336, + "estimated_input_tokens": 9084, + "estimated_eval_tokens": 18424 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a false-positive semantic neighborhood: the issues span unrelated regressions, model families, and API surfaces. I found no pair that looks like the same underlying bug/change, so all soft edges should be rejected.", + "confidence": 0.95, + "canonical_issue_reason": "No true duplicate center emerges; if one issue must be chosen, issue 34689 is a clear, well-described regression report and a reasonable representative of the cluster\u2019s broad 'model loading breakage' theme.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 34689 is the most concrete and broadly representative bug report in an otherwise heterogeneous set, but it is not a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both involve loading regressions, but one is a specific Llama 3.2 Vision breakage in 4.46.2 and the other is a v5 base_model_prefix loading change; different code paths and failure modes." + }, + { + "left": "issue:39692", + "right": "issue:45491", + "accept": false, + "reason": "Different models and different bugs: SigLIP2 docs/example errors versus Gemma3 NaN embeddings during mixed-length batching." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "OOM with Adafactor + DeepSpeed is unrelated to token embedding resize reinitialization in post_init." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "One is a weight-shape mismatch during loading, the other is an incorrect save_pretrained checkpoint for Qwen3.5; not the same underlying bug." + }, + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Different subsystems and symptoms: optimizer/deepspeed memory blowup versus model-loading regression for a specific vision model." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing in eval mode is unrelated to quantization/offloading behavior." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and stale pipeline documentation are different issues." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "PatchTSMixer post_init validation and Qwen video-input regression are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Both touch config APIs, but one is a Gemma3TextConfig inconsistency and the other is AutoModelForCausalLM unwrapping composite multimodal text_config; different fixes." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "BitNet accelerate loading with packed weights is unrelated to AutoTokenizer failing for a custom tokenizer." + }, + { + "left": "issue:41628", + "right": "issue:45372", + "accept": false, + "reason": "Both are import errors, but they concern different missing symbols and different dependency chains." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Trust-remote-code config field loss is not the same as GPTNeoX rotary_pct resetting on reload." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "FSDP CPU RAM efficient loading failure is not the same as a specific model weight-shape mismatch." + }, + { + "left": "issue:34689", + "right": "issue:36010", + "accept": false, + "reason": "Different loading errors with different root causes; one is a version regression for Llama 3.2 Vision, the other is a missing GenerationMixin import path." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior and Gemma-3 eos_token_id inconsistency are unrelated." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Dataset loading in a training script is unrelated to unrecognized model_type in config.json." + }, + { + "left": "issue:33290", + "right": "issue:36331", + "accept": false, + "reason": "OOM in DeepSpeed/Adafactor is unrelated to CustomTrainer.compute_loss signature mismatch." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 IndentationError on import is unrelated to a DeepSpeed ZeRO3 deque underflow." + }, + { + "left": "issue:43316", + "right": "issue:44877", + "accept": false, + "reason": "Both involve configs, but the specific discrepancies and affected models differ; not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Both are multimodal processor bugs, but one is Qwen3OmniMoe video processing and the other is batched apply_chat_template padding behavior in Qwen2_5_VL." + }, + { + "left": "issue:42175", + "right": "issue:43316", + "accept": false, + "reason": "Packaging/backend dependency omission is unrelated to a Gemma3TextConfig API discrepancy." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Both mention cache, but the first is missing GenerationMixin cache in chunked_prefill while the second is recurrent state reset in modular qwen3_5; different code paths." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "A dummy Conv2d implementation issue is not the same as a CUDA OOM in Sam3Video." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Mistral3 lm_head tying and Ministral-3 FlashAttention failure are distinct bugs." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 breakage affecting workflows and a janus image-generation test failure are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Processor.tokenizer/image passing regression is different from a pipeline error message about translation support." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline top_k nesting behavior is unrelated to a missing PyTorch dependency ImportError." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Label truncation in gather_for_metrics is unrelated to auxiliary-loss normalization in OLMoE/GPT Oss." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:32090", + "issue:35141", + "issue:41950", + "issue:42617", + "issue:43065", + "issue:43295", + "issue:43452", + "issue:43475", + "issue:43504", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43701", + "issue:43742", + "issue:43761", + "issue:43827", + "issue:43906", + "issue:43931", + "issue:43994", + "issue:44060", + "issue:44077", + "issue:44190", + "issue:44242", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44521", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45290", + "issue:45381", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45698" + ], + "soft_pairs": [ + "issue:41950|issue:43475", + "issue:44077|issue:44534", + "issue:16998|issue:22355", + "issue:43550|issue:43701", + "issue:44871|issue:45584", + "issue:43761|issue:44479", + "issue:43606|issue:43701", + "issue:45127|issue:45200", + "issue:43295|issue:44871", + "issue:43452|issue:44496", + "issue:42617|issue:43575", + "issue:43575|issue:43742", + "issue:44734|issue:45406", + "issue:43646|issue:44913", + "issue:43504|issue:45698", + "issue:44617|issue:45584", + "issue:43761|issue:44242", + "issue:43931|issue:44060", + "issue:45563|issue:45584", + "issue:32090|issue:35141", + "issue:44464|issue:44871", + "issue:44479|issue:44734", + "issue:43065|issue:44610", + "issue:44792|issue:44977", + "issue:44521|issue:45381", + "issue:43906|issue:45290", + "issue:43827|issue:43994", + "issue:44464|issue:45563", + "issue:44534|issue:44913", + "issue:44190|issue:44964" + ], + "prepared_review_unit_hash": "2e4e383b1b7301f3b6a9c5a6cbf4247fa6bce2ad146eaaefa5fca5a9948f2d87", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36073, + "estimated_input_tokens": 9019, + "estimated_eval_tokens": 18294 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is heterogeneous: most soft pairs share only broad subsystem or wording similarity, not the same underlying bug. No pull requests are present. Issue 44521 is the best anchor only because it is open, active, and well-scoped.", + "confidence": 0.42, + "canonical_issue_reason": "Issue 44521 is the strongest canonical anchor in this set: it is open, has the most inbound references, and describes a concrete reproducible multimodal chat-template regression.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44521 is the most triage-worthy representative issue because it is active, well-described, and likely to attract follow-up reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Different bugs: pipeline processor lookup vs missing model output attribute in SAM3 video." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "One is a post_init allowance policy change; the other is a v5 buffer initialization regression." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Unrelated topics: DeBERTa model_max_length question vs missing transformers.onnx module." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "SDPA/torch.compile failure is unrelated to resume_from_checkpoint key mismatch." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Different code paths: eos_token_id config mismatch vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "Both are regressions, but on different models and different failures; not the same bug." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch is unrelated to checkpoint key mismatch." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA merge / tied embeddings issue is different from Gemma 4 mm_token_type_ids defaults." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Processor/tokenizer regression is unrelated to eos_token_id configuration drift." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "Both concern loading, but the root causes differ: gguf_file handling vs missing model_type in config." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "A script startup failure and an OOM during tensor-parallel loading are not the same defect." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "OOM during model load does not match a key error for MobileLLM loading." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "Both hit transformers serve, but one is KV-cache indexing and the other is a missing processor attribute." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Custom model initialization breakage is not the same as GPTNeoXConfig reload reverting rotary_pct." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Legacy field handling in BEiT preset loading is unrelated to wrong custom-module selection after save_pretrained." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "CUDA OOM in Sam3Video is unrelated to Whisper empty transcription behavior." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "Hidden_states missing in CLIPVisionModel is different from missing load-balancing loss when router logits are off." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Shape mismatch on Qwen3-VL load is not the same as tied-weights warning in Qwen3-Next." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "A stale warning in paged generate is unrelated to a generation crash on empty Whisper output." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer broadcast NoneType error is unrelated to embedding reinitialization after resize." + }, + { + "left": "issue:44464", + "right": "issue:44871", + "accept": false, + "reason": "Chunked generation inconsistency with compiled forward is unrelated to eos_token_id config inconsistency." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Both are regressions, but one concerns video input processing and the other server-side KV-cache indexing." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is different from OmDet-Turbo processor/model input-size mismatch." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Failed Janus image-generation test is unrelated to Qwen3.5 flash-attention generation." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both are multimodal, but assistant mask generation and video vision_position_ids are different bugs." + }, + { + "left": "issue:43906", + "right": "issue:45290", + "accept": false, + "reason": "An isolated reproduction of an older issue is not enough to match the tool-call/no-content chat-template crash." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Documentation references to removed pipelines do not match a model-output correctness problem." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Generation inconsistency and a stale warning are different severity and different failure modes." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer junk in v5 is unrelated to GPTNeoX config reload defaults." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading in an example script is unrelated to loading a multimodal model preset." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:35141", + "issue:36246", + "issue:38175", + "issue:38617", + "issue:39692", + "issue:41628", + "issue:41720", + "issue:41950", + "issue:42490", + "issue:43065", + "issue:43295", + "issue:43335", + "issue:43388", + "issue:43479", + "issue:43540", + "issue:43550", + "issue:43644", + "issue:43650", + "issue:43976", + "issue:44016", + "issue:44077", + "issue:44079", + "issue:44112", + "issue:44186", + "issue:44190", + "issue:44220", + "issue:44462", + "issue:44496", + "issue:44734", + "issue:44737", + "issue:44792", + "issue:44855", + "issue:44913", + "issue:44933", + "issue:44938", + "issue:45127", + "issue:45245", + "issue:45276", + "issue:45357", + "issue:45362", + "issue:45375", + "issue:45542", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45803", + "issue:45907" + ], + "soft_pairs": [ + "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:44079", + "issue:35141|issue:42490", + "issue:44462|issue:45907", + "issue:43335|issue:44855", + "issue:44077|issue:44079", + "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:43335|issue:44190", + "issue:43650|issue:44220", + "issue:45276|issue:45584", + "issue:39692|issue:44190", + "issue:45245|issue:45584", + "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" + ], + "prepared_review_unit_hash": "a5a112a68f564965a0408cf3f9900ba9ba3db224c9f535914556750a49544df1", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36323, + "estimated_input_tokens": 9081, + "estimated_eval_tokens": 18418 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft pairs are mostly high-similarity false positives: they overlap in broad subsystem vocabulary (imports, configs, pipelines, CI, model loading) but describe different models, code paths, and failure modes. I would not deduplicate any of them.", + "confidence": 0.97, + "canonical_issue_reason": "If a single issue must be chosen as the representative, issue 45276 is the clearest, most concrete standalone bug in this set: it precisely describes a resize-token-embeddings/output-embeddings mismatch with a reproducible model behavior change.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45276 is the strongest issue candidate because it is specific, actionable, and tied to a concrete state inconsistency after embedding resizing; it is the most understandable representative among otherwise unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Different problems: an invalid import from image_utils vs a Qwen3.5 chat runtime crash." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Unrelated failures: auto device mapping CUDA assert vs category-count limit error." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI-ish, but one is a janus image-generation test failure and the other is a stale auto-mapping reference bug." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Completely different code paths: Python 3.13 parsing/IndentationError vs flash-attention aux-state handling." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "Different bugs: XLNet device placement in positional encoding vs a bare-except swallowing errors in Fuyu tensor conversion." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "Same model family, but distinct issues: LoRA merge with extended vocab vs incorrect visual encoder keys in save_pretrained." + }, + { + "left": "issue:43388", + "right": "issue:44079", + "accept": false, + "reason": "One is metric gathering dropping tuple labels; the other is ModelOutput field assignment with None keys." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Embedding reinitialization after resize is unrelated to the local-path trailing-slash import warning." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Both involve repository/file lookup, but tokenizer.json precedence and list_repo_files revision forwarding are different APIs and failures." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "Different models and bugs: SwitchTransformers sparse-layer creation vs DebertaV2 Python 3.13 import parsing." + }, + { + "left": "issue:44077", + "right": "issue:44079", + "accept": false, + "reason": "Unrelated: patchtsmixer post_init policy vs ModelOutput key reassignment behavior." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 import/load failure is not the same as the tensorboard-without-TensorFlow backend error." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Different surfaces entirely: torch.compile/SDPA failure in Bamba vs dataset loading in an image-classification script." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "Both are version-compatibility reports, but one targets Python 3.9/3.10 behavior and the other is a Python 3.13 IndentationError." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffer initialization junk is not the same as GPTNeoX rotary_pct not persisting on reload." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Different runtime paths: Qwen3OmniMoe video-input validation vs transformers serve KV-cache indexing." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "A stale GraniteSpeech test is not the same as broken auto_mappings references to removed configs." + }, + { + "left": "issue:44496", + "right": "issue:45803", + "accept": false, + "reason": "Model config discovery/import failure is unrelated to FuyuBatchFeature's broad exception handling." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "Different subsystems and symptoms: SwitchTransformers config creation vs local dataset loading in a training script." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "The placeholder 'ADD THE DATA' issue is not the same as an audio feature extraction bug." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Embedding-resize/output-embedding inconsistency is unrelated to Whisper empty-transcription generation after token alignment." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "Docs example/quantization failures are distinct from a local dataset loading problem." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Different errors and models: category cardinality limit vs Whisper generation on empty transcription." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "One is a video-classification pipeline bug; the other is a notebook syntax error." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Both are config issues, but they affect different multimodal configs and different fields/defaults." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Regression in processor.tokenizer/image passing is unrelated to Gemma4 token-embedding resize behavior." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and different model modules." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Different missing imports from different modules; not the same underlying bug." + }, + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "Tokenizer padding/truncation crash and serve KV-cache indexing crash are separate failures." + }, + { + "left": "issue:38175", + "right": "issue:43065", + "accept": false, + "reason": "Different models and mechanisms: SigLIP2 zero probabilities vs Sam3PixelDecoder dummy Conv2d." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:42175", + "issue:42490", + "issue:42907", + "issue:43295", + "issue:43299", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43673", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44075", + "issue:44112", + "issue:44220", + "issue:44246", + "issue:44279", + "issue:44336", + "issue:44351", + "issue:44368", + "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:45278", + "issue:45335", + "issue:45357", + "issue:45446", + "issue:45584", + "issue:45593", + "issue:45600", + "issue:45632", + "issue:45759" + ], + "soft_pairs": [ + "issue:44683|issue:45446", + "issue:42175|issue:44279", + "issue:44016|issue:44220", + "issue:44466|issue:44913", + "issue:44734|issue:45632", + "issue:43404|issue:45216", + "issue:44561|issue:45632", + "issue:44246|issue:45003", + "issue:42490|issue:45632", + "issue:43643|issue:44496", + "issue:44964|issue:45600", + "issue:44112|issue:44964", + "issue:42907|issue:43441", + "issue:43854|issue:44964", + "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", + "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:44351|issue:45278" + ], + "prepared_review_unit_hash": "0b8816fae2909264dc6cae672aa096cccdca1187a24b84a208a3bc33b3bcc904", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36350, + "estimated_input_tokens": 9088, + "estimated_eval_tokens": 18432 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are a heterogeneous set of unrelated Transformers issues; the soft pairs mostly only share broad subsystem or regression language, not the same concrete bug. No soft edge should be merged.", + "confidence": 0.97, + "canonical_issue_reason": "No true duplicate cluster emerges from these candidates. If one issue must serve as an umbrella, issue 45278 is the broadest (many import errors after upgrade), but it is still not the same underlying bug as the others.", + "canonical_pr_reason": null, + "best_issue_reason": "issue 45278 is the most general and umbrella-like issue in the set, so it is the least bad anchor for triage, though the set is too diverse to treat as one duplicate cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Different failures: compiled flex_attention on torch>=2.9 vs a wrong PyTorch version check for AuxRequest import." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "Packaging dependency problems, but one is missing TensorFlow in extras and the other is a general transformers dependency issue; not the same bug." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax error vs a feature-specific audio feature extraction issue; unrelated code paths." + }, + { + "left": "issue:44466", + "right": "issue:44913", + "accept": false, + "reason": "Both involve config/serialization behavior, but one is tied lm_head.weight saving and the other is rotary_pct not persisting on reload." + }, + { + "left": "issue:44734", + "right": "issue:45632", + "accept": false, + "reason": "Serve/KV-cache tensor indexing crash is unrelated to a trust_remote_code cache-path collision for local models." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Weight tying in Mistral3 conditional generation is unrelated to Qwen3.5 save_pretrained checkpoint corruption." + }, + { + "left": "issue:44561", + "right": "issue:45632", + "accept": false, + "reason": "Removal of a helper breaking trust_remote_code models is a different failure mode from local cache-path collisions." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Import latency during transformers import is not the same as unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:42490", + "right": "issue:45632", + "accept": false, + "reason": "Local path ending with slash import warning is different from cache path collisions between same leaf directory names." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Missing fields from trust_remote_code AutoConfig return values is not the same as an unrecognized-model/config.json model_type error." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Model loading failure for Phi-4 multimodal is unrelated to stale auto_mappings references to removed Sam3LiteText configs." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "A CI test regression on GraniteSpeech device override is unrelated to loading Phi-4 multimodal models." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized Ministral/Devstral models and FlashAttention loading failures are separate model-specific bugs." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Unit-test load failure for GLM-4.7-Flash and Phi-4 multimodal loading failure are different models and code paths." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error and AddedToken keyword collision are unrelated issues." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility breakage is not the same as a specific model loading failure on 5.1.0+." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Processor/tokenizer regression and t5gemma resize_token_embeddings bug affect different APIs and model components." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Generation cache missing during chunked_prefill is unrelated to ANSI codes emitted by loading_report." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and stale auto_mappings references are different bugs." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Qwen2.5-VL import/export failure is not the same as a nonexistent import from image_utils." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrecognized model/config error is different from PIL image processors incorrectly requiring torchvision." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to TextClassificationPipeline docs mismatch." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3VL MoE loading breakage is a different model family and failure mode from Qwen3.5 save_pretrained corruption." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "SGD optimizer arguments not being used is unrelated to a tie_word_embeddings warning in Qwen3.5 fine-tuning." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "t5gemma embedding resize bug is unrelated to Whisper empty transcription generation after token alignment." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python compatibility and dependency issues are both environment-related, but they are not the same concrete failure." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary losses with denoising off are unrelated training behaviors." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5-related regressions, but one is loading failure and the other is incorrect save_pretrained visual encoder keys." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Docs mentioning return_all_scores and a model config loading error are not the same bug." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model config loading failure is different from AutoModelForCausalLM not unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "Missing HybridCache import is a specific symbol export issue, not the broad import-error cluster described by 45278." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:33290", + "issue:38175", + "issue:39692", + "issue:41950", + "issue:42994", + "issue:43065", + "issue:43295", + "issue:43299", + "issue:43317", + "issue:43441", + "issue:43450", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44242", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44291", + "issue:44297", + "issue:44462", + "issue:44466", + "issue:44521", + "issue:44617", + "issue:44683", + "issue:44779", + "issue:44898", + "issue:45020", + "issue:45081", + "issue:45200", + "issue:45216", + "issue:45276", + "issue:45290", + "issue:45325", + "issue:45405", + "issue:45542", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45907" + ], + "soft_pairs": [ + "issue:44521|issue:45325", + "issue:41950|issue:45020", + "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:41950|issue:43650", + "issue:43295|issue:44779", + "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:45600|issue:45712", + "issue:43450|issue:44279", + "issue:43065|issue:44898", + "issue:42994|issue:43317", + "issue:44466|issue:45276", + "issue:44261|issue:44263", + "issue:43650|issue:45290" + ], + "prepared_review_unit_hash": "b49b549da53e24330d425c68000b5931dc6369d387d8d5fa41c09812b82d64cf", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36117, + "estimated_input_tokens": 9030, + "estimated_eval_tokens": 18316 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is largely a collection of unrelated Transformers issues with superficial similarity around regressions, tokenizers, multimodal models, and deepspeed. I found no strong duplicate pairs to merge; the soft matches are mostly same-subsystem, not same-bug matches.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45020 is the broadest representative report here: it describes a recent-version regression affecting model loading/remote code and best matches the cluster\u2019s overall \u2018v5 breakage\u2019 theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the most suitable anchor because it is broad, open, and discussion-heavy, while the other issues are narrower, model-specific failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "Both are multimodal, but one is a chat-template masking bug and the other is a RoPE/position-id scaling issue for Qwen2.5-VL. Different code paths and symptoms." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Video-classification processor lookup and remote-code model breakage are unrelated failures." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python version support failure and a PEFT minimum-version bump are not the same bug." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Load-balancing loss gating and Gemma4 mm_token_type_ids defaults are different model behaviors." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Different configs and different failures: default multimodal config initialization vs missing pad_token_id on Llama4Config." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they target different model families and likely different tokenizer code paths." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Processor selection for video classification is unrelated to Gemma4 device placement during forward." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "Zero probabilities in SigLIP2 and CUDA OOM in Sam3Video are unrelated issues." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "BLIP2 dtype propagation and Qwen3.5 save_pretrained serialization are distinct bugs." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Compiled flex_attention on torch 2.9 and tensorboard/TensorFlow backend installation are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Both involve docs, but one is a SigLIP2 example error and the other is a pipeline docs mismatch." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Different model families and different loading failures; not the same underlying regression." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A generic dependency issue is not the same as a specific unreleased PEFT version bump." + }, + { + "left": "issue:41950", + "right": "issue:43650", + "accept": false, + "reason": "Unrelated: video pipeline processor lookup vs an unrelated placeholder issue." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Processor/tokenizer regression in v4.57.5 is separate from Deepseek tokenizer misbehavior in v5." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "AutoTokenizer ignoring tokenizer.json and Mistral regex patch crashing are different tokenizer-loading bugs." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer class serialization mismatch is not the same as repository tokenizer.json being ignored." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Mistral tokenizer loading and list_repo_files revision forwarding are unrelated." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error and init_empty_weights argument mismatch are unrelated." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "FlashAttention failure on Ministral-3 and OCR index error in Kosmos2.5 are different model-specific bugs." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Both mention deepspeed/optimizers, but OOM with Adafactor and double optimizer.step calls are different issues." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both are repo hygiene/CI problems, but they concern different files and failure modes." + }, + { + "left": "issue:43450", + "right": "issue:44279", + "accept": false, + "reason": "Video processor batching shape bug and a dependency issue are unrelated." + }, + { + "left": "issue:43065", + "right": "issue:44898", + "accept": false, + "reason": "Different vision-model bugs: dummy Conv2d in Sam3PixelDecoder vs Perceiver non-default resolution failure." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Quantized model saving failure and dequantized load/offload failure are distinct stages and bugs." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Serialization of lm_head.weight and resize_token_embeddings not updating Gemma4 outputs are separate problems." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "MLA layernorm epsilon mismatch and GlmMoeDsaIndexer split-return handling are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:45290", + "accept": false, + "reason": "Placeholder issue and apply_chat_template crash with tool-call assistant messages 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": 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-10000-400", + "nodes": [ + "issue:41628", + "issue:42175", + "issue:42491", + "issue:43122", + "issue:43262", + "issue:43299", + "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:44242", + "issue:44246", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44361", + "issue:44373", + "issue:44521", + "issue:44589", + "issue:44617", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:44987", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45561", + "issue:45563", + "issue:45593", + "issue:45823" + ], + "soft_pairs": [ + "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:44361|issue:45593", + "issue:42175|issue:43976", + "issue:44671|issue:44779", + "issue:43784|issue:45542", + "issue:43901|issue:43994", + "issue:43122|issue:43650", + "issue:44987|issue:45542", + "issue:43824|issue:43976", + "issue:44297|issue:45397", + "issue:44188|issue:44336", + "issue:43525|issue:43881", + "issue:43525|issue:43643", + "issue:43450|issue:44617", + "issue:44589|issue:45362", + "issue:41628|issue:43976", + "issue:45561|issue:45563", + "issue:43901|issue:44246", + "issue:42491|issue:43299", + "issue:44336|issue:44373", + "issue:44242|issue:45593", + "issue:44295|issue:45397" + ], + "prepared_review_unit_hash": "89f96f0701692593d902868c841374adc4472e841bb338ecc00da64aff6446f1", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35460, + "estimated_input_tokens": 8865, + "estimated_eval_tokens": 17986 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a loose embedding match around Transformers v5-era regressions, but the paired items are on different code paths (tokenizers, model loading, chat templates, attention kernels, backend packaging, docs, etc.). I do not see any soft pair that is close enough to be treated as the same underlying bug/change.", + "confidence": 0.91, + "canonical_issue_reason": "issue:44521 is the most central user-facing bug in this set: it is broad, clearly described, open, and has the most obvious potential to attract/collect related multimodal chat-template regressions.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the best representative issue here because it is concrete, high-signal, and more likely than the others to serve as a meaningful anchor for nearby regressions.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Different failures: saving processor artifacts vs loading a model under zero3; no shared concrete code-path." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "Both mention attention backends, but one is FlashAttention in Ministral-3 and the other is compiled flex_attention on torch>=2.9; distinct bugs." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Processor save_pretrained missing files and tokenizer throughput regression are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "BigBirdTokenizer special-token registration is unrelated to D-FINE auxiliary loss logic." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Same API family, but different bug: audio sampling-rate default vs multimodal assistant mask generation." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer loading regression and ESM2 model breakage are different issues." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error is not the same as Gemma4 device-map forward mismatch." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Generation cache/chunked_prefill failure and CLIPVision hidden_states regression affect different modules and behaviors." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer task AttributeError and D-FINE auxiliary loss omission are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Backend packaging/install issue and Python version compatibility issue are different root causes." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Different model families and different symptoms: CamemBERT MLM predictions vs Deepseek tokenization." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Both are import/backend-related, but one is a missing nn import in sentence-transformers integration and the other is a missing TensorFlow backend when only tensorboard is installed." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Docs mismatch about return_all_scores is not the same as incorrect SigLIP2 inference outputs." + }, + { + "left": "issue:43122", + "right": "issue:43650", + "accept": false, + "reason": "Tokenizer output drift and a nonspecific 'ADD THE DATA' issue do not identify the same bug." + }, + { + "left": "issue:44987", + "right": "issue:45542", + "accept": false, + "reason": "Model loading failure for a specific repo and TensorFlow backend packaging issue are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Qwen2_5_VL import failure and Python 3.9/3.10 compatibility are different failure modes." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer config serialization mismatch and Gemma4 loading under zero3 are different code paths." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to ANSI codes in loading_report." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Llama4Config missing pad_token_id and glm-4v-9b loading failure are model-specific and not the same bug." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "One is a specific config attribute absence; the other is a generic trust_remote_code field-loss issue. Similar theme, different defect." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Incorrect batched video processor shapes and Sam3Video CUDA OOM are not the same underlying problem." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage lookup failure and Qwen3.5 chat crash are different runtime errors." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "Importing AutoImageProcessor and Python version support are unrelated." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "pytest-xdist file race and stale generate() warning are unrelated test/runtime issues." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Documentation note about return_all_scores and slow import behavior are different concerns." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MoE, but one is LoRA compatibility across versions and the other is base-model loading regression; too different to merge as one fix." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI output formatting and a wrong docstring are not the same bug." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Load-balancing loss routing logic and D-FINE auxiliary-loss behavior are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Buffer registration/readback error and Gemma4 zero3 loading failure 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": 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-10000-400", + "nodes": [ + "issue:36246", + "issue:41950", + "issue:42757", + "issue:42886", + "issue:43065", + "issue:43232", + "issue:43257", + "issue:43262", + "issue:43352", + "issue:43408", + "issue:43526", + "issue:43531", + "issue:43596", + "issue:43673", + "issue:43901", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44336", + "issue:44351", + "issue:44485", + "issue:44496", + "issue:44561", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:44877", + "issue:44987", + "issue:44995", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45229", + "issue:45250", + "issue:45292", + "issue:45362", + "issue:45375", + "issue:45459", + "issue:45561", + "issue:45584", + "issue:45701", + "issue:45759" + ], + "soft_pairs": [ + "issue:41950|issue:44220", + "issue:44246|issue:44496", + "issue:43531|issue:45362", + "issue:36246|issue:42757", + "issue:43352|issue:45250", + "issue:36246|issue:44351", + "issue:44336|issue:44485", + "issue:43526|issue:45003", + "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:42886|issue:45229", + "issue:43065|issue:43408", + "issue:44987|issue:45459", + "issue:44496|issue:45003", + "issue:45375|issue:45759", + "issue:43257|issue:45137", + "issue:43262|issue:44246", + "issue:44995|issue:45200", + "issue:44060|issue:44077" + ], + "prepared_review_unit_hash": "70abf0262e597b254f4f0fd35406be8df6929df57bd75646db8323e6ed6ffa78", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 35958, + "estimated_input_tokens": 8990, + "estimated_eval_tokens": 18236 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft pair looks like the same underlying bug or change. The set is heterogeneous across unrelated areas (pipelines, tokenizer/config loading, generation, DeepSpeed, and model-specific issues), so none should be deduplicated.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue fits this cluster: the candidates cover unrelated defects and no pair looks mergeable as the same bug.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue for deduplication; the list is too mixed to anchor a canonical representative.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Different subsystems and symptoms: video-classification pipeline processor lookup vs audio fbank feature extraction." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import performance issue is unrelated to model config parsing failure for an unrecognized model." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Both involve Qwen, but the bugs differ: sliding-window handling vs chat crash on Qwen3.5-35B." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Distinct import errors in different packages: Qwen2.5-VL image processor vs huggingface_hub offline helper." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "One is a model-specific Flash Attention support error; the other is only a broad feature title with no matching code path shown." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Different missing imports from different modules; no shared code-path or fix." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI color output in loading reports is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BEiT image processor label reduction bug is unrelated to unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe module lookup and xdist file race are separate testing/runtime problems." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Chunked-prefill cache bug is unrelated to removal of is_torch_fx_available breaking trust_remote_code models." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import slowdown has no concrete relation to Gemma eos_token_id configuration mismatch." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "Load-balancing loss omission and LWDETR AMP crash are different MoE vs vision-loss bugs." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaulting are unrelated model-specific issues." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Position_ids buffer access error is unrelated to the reported tokenization slowdown." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Qwen2.5-VL import failure and DeepSpeed ZeRO3 deque IndexError are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Strict config loading for granite_speech is not the same as missing Qwen3_5MoeVisionConfig field handling." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer regression and resize_token_embeddings output embedding mismatch are different bugs." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize behavior is unrelated to Whisper empty-transcription generation failure." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch for return_all_scores is unrelated to sys.modules access." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "resize_token_embeddings/output embeddings bug is not the same as version-dependent tokenization changes." + }, + { + "left": "issue:43232", + "right": "issue:44336", + "accept": false, + "reason": "Generation kwargs update after sync_gpus is unrelated to ANSI codes in loading_report." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading bug is unrelated to Gemma4 multi-GPU OOM." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder and sam3_video/sam3_tracker config mismatch are different SAM3 issues." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Model loading failure for physical-intelligence/fast is unrelated to tokenizer exception masking when protobuf is absent." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model/config error is unrelated to unsafe sys.modules access." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Missing config field in Qwen3_5MoeVisionConfig is not the same as AutoModelForCausalLM unwrapping text_config." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed and ZeRO3 deque popping are different failure modes." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sample-rate default bug is unrelated to import performance." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa stale indexer cache crash does not match Gemma4 token-type-id defaults." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Qwen3.5 tied-weights warning and patchtsmixer optional post_init are unrelated model/config 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": 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-10000-400", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:41628", + "issue:42371", + "issue:42994", + "issue:43232", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43540", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43673", + "issue:43931", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44361", + "issue:44368", + "issue:44466", + "issue:44623", + "issue:44661", + "issue:44734", + "issue:44749", + "issue:44869", + "issue:44933", + "issue:44977", + "issue:44995", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45137", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45335", + "issue:45356", + "issue:45375", + "issue:45381", + "issue:45397", + "issue:45459", + "issue:45478", + "issue:45563", + "issue:45588", + "issue:45600", + "issue:45656" + ], + "soft_pairs": [ + "issue:43475|issue:43540", + "issue:43653|issue:44162", + "issue:44466|issue:45335", + "issue:45397|issue:45656", + "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:43232|issue:43673", + "issue:44297|issue:44623", + "issue:43673|issue:45278", + "issue:43646|issue:44162", + "issue:44361|issue:44749", + "issue:44623|issue:45656", + "issue:44749|issue:45656", + "issue:44749|issue:45310", + "issue:28282|issue:30990", + "issue:42994|issue:43482", + "issue:45230|issue:45397", + "issue:42371|issue:45092", + "issue:43475|issue:43643", + "issue:43475|issue:45381", + "issue:44038|issue:44623", + "issue:41628|issue:44933", + "issue:45081|issue:45103", + "issue:43475|issue:44661", + "issue:42994|issue:43525", + "issue:45375|issue:45478", + "issue:43232|issue:44188" + ], + "prepared_review_unit_hash": "589c4b661defa5488cb3fea9cb22746fb1831c6b1d232049676fe27c1890884f", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37588, + "estimated_input_tokens": 9397, + "estimated_eval_tokens": 19050 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly a grab bag of unrelated regression reports across tokenizers, generation, model loading, deepspeed, vision, and import/export paths. I did not find any soft pair that looks like the same concrete bug fix target, so no duplicate merges are warranted.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45278 is the broadest current umbrella report in the set: it targets upgrade-related import regressions across many cases and is open, making it the best central triage anchor.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best single issue to keep as the representative hub because it is broad, current, and likely to accumulate related import-regression reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "Different models and failure modes: SAM 3 video missing an encoder output attribute vs Qwen3OmniMoe video-input processing ValueError." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "One is a BigBirdTokenizer special-token registration bug; the other is an ESM2 breakage report affecting workflows, not the same code path." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Both involve weight/embedding serialization, but one is tied lm_head serialization by device and the other is resize_token_embeddings not updating decoder.embed_tokens; different bugs." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Both mention DeepSpeed, but one is a gemma-4 from_pretrained/ZeRO3 issue and the other is double optimizer stepping; unrelated code paths." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "A second-forward stale indexer cache crash is not the same as a flash-attention crash on s_aux=None." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "One is a tie_word_embeddings warning during fine-tuning; the other is a stale num_return_sequences warning in paged generate()." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "A warning about tie_word_embeddings is unrelated to Qwen3.5 generation failures under flash-attention." + }, + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Both touch KV/cache behavior, but one is missing GenerationMixin cache in chunked_prefill and the other is wrong tensor indexing during /v1/responses continuation." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque underflow is unrelated to stale auto_mappings references to removed Sam3LiteText configs." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch is unrelated to a config warning about tie_word_embeddings." + }, + { + "left": "issue:44933", + "right": "issue:45459", + "accept": false, + "reason": "A missing import from image_utils is not the same as protobuf decode errors being masked by except import_protobuf_decode_error()." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper trailing-replacement-character timestamp decoding and Kimi-K2.5 codec/fix_mistral_regex regression are different tokenizer issues." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Both are generation-related, but one is sync_gpus/update_model_kwargs logic and the other is v5 chunked_prefill cache availability." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "tokenizer.save_pretrained writing the wrong tokenizer_class is a different save-path bug from processor.save_pretrained missing files." + }, + { + "left": "issue:43673", + "right": "issue:45278", + "accept": false, + "reason": "Chunked_prefill cache handling is not the same as broad import errors after upgrading from 4.57.0 to 5.5.0." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Custom model initialization breakage in v5 is too broad to equate with an ESM2-specific workflow regression." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "MLukeTokenizer AttributeError on tasks is unrelated to performance slowdown in filtering after upgrading Transformers." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Missing files from processor.save_pretrained is unrelated to DeepSpeed causing the optimizer step to run twice." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "A performance regression in data filtering and a DeepSpeed double-step bug are not the same underlying issue." + }, + { + "left": "issue:44749", + "right": "issue:45310", + "accept": false, + "reason": "Slow filtering after upgrade is unrelated to Qwen3.5 Moe from_pretrained errors." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "PyTorch-missing ImportError and Sentence Transformers hanging on load are different environment/loading problems." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure is not the same as Qwen2.5-GGUF loading failure in v5." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "A placeholder bug report does not identify the same from_pretrained/ZeRO3 issue as gemma-4." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 API-setting guidance is unrelated to old InternVL2 checkpoints breaking under meta initialization." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM 3 video encoder output missing an attribute is unrelated to AutoConfig returning missing fields under trust_remote_code=True." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "Different vision/video regressions in different models: SAM 3 output attribute error vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe from_pretrained breakage and processor.save_pretrained missing files are separate bugs." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "An AutoImageProcessor import error and a missing image_utils import are related only at a broad import level, not the same concrete bug." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Mistral regex patch crashing on tokenizer backend access is unrelated to auto_docstring crashing on future annotations." + }, + { + "left": "issue:43475", + "right": "issue:44661", + "accept": false, + "reason": "SAM 3 vision encoder output and add-new-model-like failing inside TOKENIZER_MAPPING_NAMES are different subsystems and failures." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized model saving and Llama4Config missing pad_token_id are unrelated." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Missing deepstack_visual_indexes in a vision config is not the same as Qwen3.5 Moe from_pretrained errors." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "GenerationMixin cache handling after sync_gpus is unrelated to diverging attention kernels under torch.compile." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:30990", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:42994", + "issue:43232", + "issue:43278", + "issue:43377", + "issue:43479", + "issue:43504", + "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", + "issue:44351", + "issue:44373", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44792", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:44964", + "issue:45103", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45375", + "issue:45561", + "issue:45712", + "issue:45753", + "issue:45812" + ], + "soft_pairs": [ + "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:44164", + "issue:42175|issue:45712", + "issue:42222|issue:45200", + "issue:43232|issue:44792", + "issue:44038|issue:45310", + "issue:43377|issue:44242", + "issue:43526|issue:45561", + "issue:43504|issue:45561", + "issue:44496|issue:45561", + "issue:44964|issue:45137", + "issue:42175|issue:44933", + "issue:42898|issue:44568", + "issue:30990|issue:35141", + "issue:43479|issue:44869", + "issue:43976|issue:44933", + "issue:43479|issue:44079", + "issue:44188|issue:44373", + "issue:43784|issue:43824", + "issue:44623|issue:45310", + "issue:42898|issue:45812", + "issue:43278|issue:45137", + "issue:44734|issue:45103" + ], + "prepared_review_unit_hash": "730eb66fc0268471083c5f17f988d6e7d699dda0f25f9a5955360f320d9e2a11", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36787, + "estimated_input_tokens": 9197, + "estimated_eval_tokens": 18650 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heavily over-broadened: most items are unrelated issues that only share superficial subsystem or model-family similarity. The only clear duplicate pair is the Qwen3ForCausalLM VRAM leak report.", + "confidence": 0.88, + "canonical_issue_reason": "issue:42673 is the clearest representative of the only true duplicate pair, with a detailed report and higher discussion activity than its twin.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the strongest global issue candidate because it matches an exact duplicate, has the fuller description, and appears earlier than the matching report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Exact same title and same underlying bug: Qwen3ForCausalLM VRAM leak in multiple dataloader threads." + }, + { + "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; too different to merge." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Different failures: quantization offloading vs incorrect outputs under pipeline parallelism." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and different code paths." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Accuracy regression in FineGrainedFP8 is not the same bug as FP8 save_pretrained failure." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output corruption and Qwen3_5 image infinite loop are unrelated symptoms." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "Saving a quantized MoE model and loading Qwen3.5 MoE with from_pretrained are different concrete problems." + }, + { + "left": "issue:43673", + "right": "issue:44188", + "accept": false, + "reason": "Generation cache handling during chunked_prefill is unrelated to attention-kernel divergence under torch.compile." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Different missing imports: GenerationMixin versus AutoImageProcessor." + }, + { + "left": "issue:42994", + "right": "issue:44164", + "accept": false, + "reason": "Quantized model saving and extra_state save/from_pretrained handling are different serialization bugs." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Packaging/backend dependency issue is not the same as leftover dummy classes leaking into dir(transformers)." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "VitPose breakage and Gemma 4 mm_token_type_ids behavior are unrelated model-specific issues." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs update after sync_gpus is unrelated to a janus image-generation test failure." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Different model families and different load failures; too broad to treat as one bug." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "Missing padding-mask support in MIMI encoder is unrelated to MoE load-balancing loss computation." + }, + { + "left": "issue:43526", + "right": "issue:45561", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to the xdist captured_info.txt race." + }, + { + "left": "issue:43504", + "right": "issue:45561", + "accept": false, + "reason": "Legacy-field pretrained loading failure is unrelated to the testing-utils race." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Unrecognized model/config loading error is unrelated to the xdist file race." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal loading and DeepSpeed ZeRO3 deque error are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Backend dependency installation and nonexistent image_utils import are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenization regressions, but they affect different behaviors and models; not the same bug." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Sentence Transformers loading hang and embedding reinitialization in post_init are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config defaults and Whisper timestamp decode crash are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version compatibility and missing import from image_utils are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config initialization bug and ModelOutput key assignment bug are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel branching under torch.compile is unrelated to a docstring issue." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Missing nn import in sentence-transformers and missing Qwen2_5_VLForConditionalGeneration import are different import problems." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "processor.save_pretrained missing files is unrelated to Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Tokenization behavior change and Granite token-ID regression are different concrete regressions." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift in evaluation is unrelated to the ZeRO3 deque error." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "KV-cache continuation crash in transformers serve is unrelated to the auto_docstring annotation crash." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative. The one accepted pair, issue:42673 vs issue:43704, is supported by identical titles plus the fuller/earlier report on 42673. The remaining soft-edge rejections stay cautious and do not overclaim duplication from superficial 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": 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-10000-400", + "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:43479", + "issue:43644", + "issue:43723", + "issue:43874", + "issue:44060", + "issue:44190", + "issue:44206", + "issue:44222", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44360", + "issue:44373", + "issue:44448", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44568", + "issue:44623", + "issue:44779", + "issue:44933", + "issue:44977", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45310", + "issue:45356", + "issue:45357", + "issue:45405", + "issue:45406", + "issue:45656", + "issue:45712", + "issue:45834" + ], + "soft_pairs": [ + "issue:30333|issue:33290", + "issue:43723|issue:44568", + "issue:43874|issue:45092", + "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:44206|issue:45103", + "issue:42831|issue:44623", + "issue:41628|issue:42757", + "issue:22355|issue:28282", + "issue:30333|issue:36331", + "issue:44297|issue:45310", + "issue:30333|issue:34689", + "issue:43232|issue:45834", + "issue:44222|issue:44484", + "issue:43723|issue:44977", + "issue:28282|issue:30333", + "issue:44263|issue:45092", + "issue:30333|issue:35141", + "issue:44483|issue:45406", + "issue:44263|issue:44360", + "issue:44373|issue:44485", + "issue:44295|issue:44623", + "issue:44514|issue:45103", + "issue:44261|issue:45092", + "issue:45092|issue:45357" + ], + "prepared_review_unit_hash": "bf533617fb1ce050b8172560b7b6a38b804bd6acfb10c63eff1c7e7c236f00d6", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38105, + "estimated_input_tokens": 9527, + "estimated_eval_tokens": 19310 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are only loosely related by broad Transformers themes (serialization/save_pretrained, tokenizer regressions, multimodal/model-loading bugs). None of the soft pairs look like the same underlying bug or change, so I rejected all of them. No PRs are present.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 44222 is the strongest concrete bug report in the loose save_pretrained/serialization subcluster, with the clearest actionable failure mode among the candidates.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44222 is the best overall anchor here: it is a specific, bug-shaped report rather than a question or broad symptom, even though the surrounding items are not true duplicates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow failure reporting is unrelated to the deepspeed/Adafactor OOM." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Different tokenizer regressions: load failure vs add_special_tokens/BOS-EOS behavior." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Missing image-patch API and remote-code meta-init incompatibility are different bugs." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "v5 tokenization-space behavior change and Kimi codec handling/warning are distinct tokenizer issues." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Incorrect tied-weights warning is not the same as a missing _tp_plan for tensor parallelism." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both involve tokenizers, but one is a v5 spacing change and the other is DeepSeek output corruption." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "A released-version dependency bump and leftover dummy classes are unrelated maintenance problems." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Both are version regressions, but they affect different models and symptoms." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Buffer-reading error and double optimizer-step behavior are unrelated code paths." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Missing image_utils import and dummy_pt_objects leftovers are different import/export issues." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "Both mention save_pretrained, but one is FP8 MoE saving and the other is tokenizer metadata mismatch." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata and duplicate optimizer steps are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Example-script multi-label inference failure and local dataset loading failure are different problems." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Config initialization with None and junk-filled non-persistent buffers are different bugs." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Both are serialization-related, but model save_pretrained FP8 behavior is not the same as processor.save_pretrained missing files." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "Unsupported feature argument in a feature extractor is unrelated to auto_docstring annotation crashes." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FineGrainedFP8 accuracy problems are unrelated to missing processor save files." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Different missing imports from different packages/modules." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "transformers.onnx import failure and missing PyTorch are separate import-environment issues." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "MLflow not reported on failure is unrelated to the CustomTrainer compute_loss signature breakage." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save metadata mismatch and Qwen3.5 from_pretrained failure are different load/save bugs." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "Training-job reporting and Llama 3.2 Vision model loading regression are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs update after sync_gpus is not the same as an OCR long-input index error." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "A concrete save_pretrained bug and a question about max_shard_size are not duplicates." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading in v5 and Qwen3.5 flash-attention generation failure are different issues." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "Missing PyTorch during AutoModel import and MLflow failure reporting are unrelated." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return handling and InternVL2 meta-init incompatibility are different model bugs." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "MLflow reporting on failure is unrelated to embedding reinitialization in post_init." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat-completions request rejection and Gemma4Processor missing _tokenizer are different API/processor failures." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Same DSA indexer area, but one is a split-return bug and the other is a missing ReLU concern." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "A wrong docstring and a RoPE implementation discussion are not the same bug." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading a registered buffer and processor save_pretrained file loss are unrelated." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "Chat-template batching crash and auto_docstring annotation crash are different code paths." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "MLA layernorm epsilon precision mismatch and InternVL2 meta-init incompatibility are different bugs." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Old InternVL2 meta-init issues and Qwen3.5 visual-key save_pretrained regressions 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": 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-10000-400", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:30990", + "issue:33357", + "issue:38175", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43122", + "issue:43232", + "issue:43479", + "issue:43576", + "issue:43723", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:43976", + "issue:44038", + "issue:44060", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44442", + "issue:44488", + "issue:44514", + "issue:44521", + "issue:44568", + "issue:44704", + "issue:44743", + "issue:44871", + "issue:44913", + "issue:44945", + "issue:44991", + "issue:45278", + "issue:45290", + "issue:45310", + "issue:45593", + "issue:45600", + "issue:45753" + ], + "soft_pairs": [ + "issue:43723|issue:44871", + "issue:30990|issue:38175", + "issue:43723|issue:44991", + "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:43232|issue:44743", + "issue:44295|issue:44297", + "issue:44704|issue:45290", + "issue:42907|issue:44945", + "issue:41628|issue:45278", + "issue:43122|issue:44514", + "issue:43976|issue:44945", + "issue:43784|issue:45600", + "issue:39401|issue:44488", + "issue:44261|issue:44373" + ], + "prepared_review_unit_hash": "465037515db1d7074a19085e82715cb2cbb6e677b68bee01aac2d202febe3f65", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36842, + "estimated_input_tokens": 9211, + "estimated_eval_tokens": 18678 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: most items are unrelated tokenizer/import/config/regression reports with only superficial similarity, so there is no strong duplicate core. The broadest representative is the v5 tokenizer-loading regression issue.", + "confidence": 0.41, + "canonical_issue_reason": "Issue 43723 is the broadest, most central tokenizer-loading regression in the set and overlaps the most with other tokenizer-related reports, though the cluster is still quite mixed.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43723 is the best single representative because it is broad, user-facing, and closest to the common tokenizer-loading theme seen in several nearby candidates.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Both are tokenizer-related, but one is a generic v5 loading failure and the other is a Gemma-3 EOS config mismatch; different bugs." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Completely different symptoms and code paths: Sentence Transformers loading hang vs zero probabilities in a SigLIP2 model." + }, + { + "left": "issue:43723", + "right": "issue:44991", + "accept": false, + "reason": "Both mention tokenizer loading in v5, but they are model-specific failures and do not look like the same underlying bug." + }, + { + "left": "issue:43723", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer implementations and failure modes; no evidence they share the same concrete bug." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different subsystems." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Unrelated config bugs: multimodal default config initialization vs GPTNeoX rotary_pct persistence." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "Accuracy issue in FineGrainedFP8 vs a Qwen3.5 MoE from_pretrained error; no shared code-path." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "One is a tensor-split indexing bug, the other is a docstring issue; not the same change or bug." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer-related, but one is offset mapping in Qwen3 and the other is missing BOS/EOS insertion for mdeberta-v3." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "Different problems: numeric accuracy with FP8 vs Qwen3-VL-Moe loading/compatibility." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "One specific missing import from huggingface_hub and one broad import-error report; too general to treat as the same bug." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading vs quantized model saving failure are unrelated." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation state update bug vs D-FINE auxiliary-loss behavior; different areas entirely." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Different model-specific architecture/config issues: RoPE layer handling vs tied-weights warning." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Broken env CLI command vs import-time NameError in sentence-transformers; not the same bug." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue and Qwen3_5 infinite loop are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Both involve tokenizer/v5 behavior, but one is offline cache loading and the other is tokenizer decoder typing; different failures." + }, + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization/chat templates, but one is version-dependent tokenization drift and the other is multimodal assistant mask generation." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 settings warning vs a torch.split indexing issue; unrelated." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "No module import error for transformers.onnx vs extra-space tokenizer behavior; unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Both are from_pretrained-related, but one is TOKENIZER_MAPPING_NAMES handling and the other is AutoProcessor kwargs forwarding." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Loading hang in sentence-transformers vs MacOS bus error in a CLIP model; not the same bug." + }, + { + "left": "issue:43232", + "right": "issue:44743", + "accept": false, + "reason": "Both touch generation/cache behavior, but the concrete failures differ: sync_gpus kwargs update vs recurrent-state reset in modular_qwen3_5." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Reading a bufferized position_ids tensor vs tokenizer_class mismatch in saved config; unrelated." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "AutoProcessor kwargs propagation vs chat_template crash with tool-call assistant messages; different code paths." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Dequantized model saving failure vs incorrect output under pipeline parallelism; unrelated." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Specific missing AutoImageProcessor import vs broad import-error report; not the same concrete issue." + }, + { + "left": "issue:43122", + "right": "issue:44514", + "accept": false, + "reason": "Both are chat/tokenization issues, but one is version tokenization drift and the other is a batched apply_chat_template crash." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility problem vs pipeline-parallelism output bug; unrelated." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import NameError in sentence-transformers vs CI break from removed auto_mappings config references; different failures." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Both are tokenizer loading-related, but the affected models and failure modes are different." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "Precision/config issue in MLA q_a_layernorm vs a docstring correction; not duplicates." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both involve apply_chat_template, but one is a batched-padding crash and the other is zero assistant masks for multimodal inputs; distinct bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:30990", + "issue:36296", + "issue:38617", + "issue:39401", + "issue:42371", + "issue:42673", + "issue:42757", + "issue:42907", + "issue:43064", + "issue:43066", + "issue:43116", + "issue:43122", + "issue:43278", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43493", + "issue:43673", + "issue:43784", + "issue:43824", + "issue:43874", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44485", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:44977", + "issue:45072", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45446", + "issue:45478", + "issue:45753" + ], + "soft_pairs": [ + "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:29127|issue:30990", + "issue:44038|issue:44297", + "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:43493|issue:45753", + "issue:43278|issue:44945", + "issue:42907|issue:44222", + "issue:43122|issue:44977", + "issue:43992|issue:44704", + "issue:43784|issue:45446", + "issue:43066|issue:44779", + "issue:38617|issue:45278", + "issue:42673|issue:45072", + "issue:43937|issue:44485", + "issue:22355|issue:29127", + "issue:42371|issue:43874", + "issue:43064|issue:43673", + "issue:43976|issue:45446", + "issue:43116|issue:43493" + ], + "prepared_review_unit_hash": "57afcd8047dc4ffa57986e6bc57645097e1888a68b534d60563f8a93fe19c1f8", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36958, + "estimated_input_tokens": 9240, + "estimated_eval_tokens": 18736 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: the soft pairs share broad themes like imports, tokenization, or Qwen/SAM/GLM models, but they do not appear to be the same concrete bug or change. I would keep them separate; no PRs are present.", + "confidence": 0.71, + "canonical_issue_reason": "issue:45278 is the broadest umbrella for the import-regression subset (many import errors after updating to 5.5.0), making it the best representative issue in this mixed cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 best captures the most central recurring theme here: upgrade-related import failures across multiple reports. It is still not a duplicate of the others, just the most useful anchor issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Same model family, but different bugs: a model-type warning vs a missing attribute on Sam3VisionEncoderOutput." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Different failures: dequantized save_pretrained vs GLM-5 GenerationConfig validation." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "Unrelated: DSA indexer ReLU discussion vs GLM-5 RoPE implementation." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Different code paths: Qwen3.5 num_labels propagation vs flash_attention_3 degenerate training." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Tokenizer offset_mapping bug vs multimodal apply_chat_template assistant mask bug." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Example-script multi-label output bug vs Gemma 4 mm_token_type_ids defaulting issue." + }, + { + "left": "issue:29127", + "right": "issue:30990", + "accept": false, + "reason": "LayoutLMv3 error-message clarity vs Sentence Transformers loading hang; not the same defect." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe regression vs tokenizer save_pretrained metadata mismatch." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Different model-loading/saving issues: dequantized save failure vs Phi-4 multimodal load failure." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they concern different models and likely different root causes." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Different bugs: Qwen3-VL-Moe regression vs position_ids buffer access." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training bug vs classification example returning empty results." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Different loading/saving paths: dequantized save_pretrained vs gguf_file loading failure." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Both import errors, but for different symbols and causes; not the same underlying bug." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Same broad MoE area, but Qwen3-VL-Moe regression and FP8 save_pretrained are different issues." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 regressions, but one is save_pretrained checkpoint corruption and the other is from_pretrained failure." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Distinct import failures from different packages/symbols, not a shared root cause." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Different model families and symptoms: SigLIP2 discrepancy vs Qwen3_5 image-specific infinite loop." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Dtype mismatch between train/eval vs incorrect output under pipeline parallelism." + }, + { + "left": "issue:42907", + "right": "issue:44222", + "accept": false, + "reason": "Both mention save_pretrained, but dequantized save and FP8 MoE save are different bugs." + }, + { + "left": "issue:43122", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer behavior regression vs generation failure with flash-attention; different concrete problems." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Separate loading bugs: missing embed_tokens.weight vs cached_file kwargs not forwarded." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "Import-related, but one is a NameError in sentence-transformers and the other is a version check bug in flex_attention." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Different tokenizer regressions with different symptoms and likely different fixes." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "45278 is a broad import-regression report, but 38617 is a specific missing symbol; not the same concrete bug." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in dataloader threads vs bfloat16 dtype mismatch in inference." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "GenerationConfig validation issue vs RoPE implementation discussion." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "No module named transformers.onnx vs LayoutLMv3 box-information error message; unrelated." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 API/settings issue vs missing image-patch count method in GLM46VImageProcessorFast." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Different training/runtime failures: FSDP2 optimizer-state bug vs missing GenerationMixin cache in chunked prefill." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python-version compatibility issue vs incorrect PyTorch version check for AuxRequest import." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Example-script classification bug vs SigLIP2 implementation discrepancy." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:33453", + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42222", + "issue:42371", + "issue:42490", + "issue:42617", + "issue:42673", + "issue:43116", + "issue:43408", + "issue:43421", + "issue:43441", + "issue:43450", + "issue:43582", + "issue:43742", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:43937", + "issue:43986", + "issue:44038", + "issue:44075", + "issue:44117", + "issue:44261", + "issue:44291", + "issue:44410", + "issue:44484", + "issue:44779", + "issue:44829", + "issue:44933", + "issue:44936", + "issue:44977", + "issue:45071", + "issue:45081", + "issue:45083", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45357", + "issue:45405", + "issue:45446", + "issue:45478", + "issue:45542", + "issue:45561", + "issue:45563", + "issue:45588" + ], + "soft_pairs": [ + "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:43827|issue:45446", + "issue:39401|issue:44779", + "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:44484|issue:45310", + "issue:44829|issue:45216", + "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" + ], + "prepared_review_unit_hash": "5d0bdbcdb34ece144caccd9921d5e6a848e590a8662f160eaa2d39da66a95cb7", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36465, + "estimated_input_tokens": 9117, + "estimated_eval_tokens": 18490 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly separate regression reports across different models, tokenizer paths, and runtime subsystems. I do not see any soft pair that clearly describes the same underlying bug, so I would not merge any of them as duplicates.", + "confidence": 0.39, + "canonical_issue_reason": "issue:45278 is the broadest umbrella report in the set: a current open issue about widespread import failures after upgrading to v5.5.0, which best matches the cluster\u2019s import-regression theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most suitable representative issue because it is open, broad, and could subsume related import-error regressions better than the highly specific model- or API-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "Different failures: VitPose model loading breakage vs Ministral-3 FlashAttention incompatibility." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Unrelated training bugs: SGD args ignored vs trainer.evaluate() failing after train()." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Different concerns: multi-label example returning empty results vs pipeline() error text about unsupported translation tasks." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but different code paths: save_pretrained key layout vs from_pretrained loading error." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen/MoE, but the model/version and failure mode differ; not the same concrete bug." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Completely different subsystems: qwen3_omni_moe helper logic vs flex_attention import version check." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "GenerationConfig validation error is unrelated to the stale warning for num_return_sequences." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer mapping None bug vs flash_attention crash on s_aux=None are unrelated." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Both tokenizer-related, but one is loading regression and the other is offset_mapping correctness for Qwen3." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs referencing pipeline() vs a PyTorch version check bug are unrelated." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Different tokenizer regressions affecting different models and symptoms." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "Invalid GenerationConfig vs flash-attention generation problems are different bugs." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "VRAM leak in dataloader threads vs an isolated reproduction issue; no shared concrete defect." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "init_empty_weights argument handling vs PretrainedConfig type checking are separate code-path regressions." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Video processor batch-shape bug vs missing torchvision crash when using AutoProcessor are different failures." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "PEFT version bump issue and flex_attention import check are unrelated." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "A testing_utils bug and a pytest-xdist race in patched testing utils are not the same issue." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer mapping fallback bug vs Mistral regex patch crash are different tokenizer paths." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different symbols and different underlying regressions." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Qwen3 offset_mapping bug vs runtime special-token/post-processor update issue are not the same defect." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing projections in qwen3next is unrelated to Qwen3.5 from_pretrained loading failure." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "save_pretrained shard-size question vs from_pretrained error are not duplicates." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "FlashAttention3 training degeneration is unrelated to incorrect checkpoint saving." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API guidance issue vs MLA rms_norm_eps precision bug are different topics." + }, + { + "left": "issue:42673", + "right": "issue:43582", + "accept": false, + "reason": "VRAM leak in threads vs Apple Silicon TypeError in caching_allocator_warmup are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "Model-type warning for sam3 vs key error loading MobileLLM are unrelated." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "3d_parallel.py execution failure and a testing_utils bug do not share the same code path." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "Related only by broad import-error theme; 45278 is an umbrella report, not the same concrete bug as the image_utils import issue." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "VitPose breakage and a TensorFlow backend error from tensorboard-only installs are unrelated." + }, + { + "left": "issue:42490", + "right": "issue:42673", + "accept": false, + "reason": "Trailing-slash local model path warning vs VRAM leak in dataloader threads 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": 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-10000-400", + "nodes": [ + "issue:38617", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:43012", + "issue:43066", + "issue:43097", + "issue:43122", + "issue:43335", + "issue:43408", + "issue:43441", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43673", + "issue:43874", + "issue:43931", + "issue:43937", + "issue:43976", + "issue:43986", + "issue:44038", + "issue:44075", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44263", + "issue:44368", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44496", + "issue:44829", + "issue:44987", + "issue:45003", + "issue:45071", + "issue:45245", + "issue:45341", + "issue:45372", + "issue:45459", + "issue:45464", + "issue:45478", + "issue:45542", + "issue:45656" + ], + "soft_pairs": [ + "issue:43937|issue:44496", + "issue:42175|issue:42222", + "issue:44483|issue:45464", + "issue:38617|issue:44220", + "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", + "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" + ], + "prepared_review_unit_hash": "038bff0d50e6d97c538201a87f7830d2f1570620b2515b4ef0e337d42638bd91", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 37997, + "estimated_input_tokens": 9500, + "estimated_eval_tokens": 19256 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is noisy and mostly contains unrelated issue reports that only look similar at the title level. I found no soft pair that clearly describes the same underlying bug/change, so I would not merge any of them.", + "confidence": 0.57, + "canonical_issue_reason": "Issue 45071 is the broadest, most central Transformers-v5 regression in the set and best represents the cluster\u2019s general compatibility-breakage theme.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45071 is the most reusable representative: it is a cross-cutting v5 regression, whereas most others are narrower model- or feature-specific failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GLM generation-config validation vs model-type/config loading failure; different bugs." + }, + { + "left": "issue:42175", + "right": "issue:42222", + "accept": false, + "reason": "Backend installation packaging issue vs VitPose model breakage; unrelated." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "Both mention chat/completions, but one is a generic request rejection and the other is streaming inference on a specific model." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "ImportError for a config symbol vs audio feature extraction issue; no concrete overlap." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 settings warning is unrelated to FineGrainedFP8 accuracy." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Tokenizer decoder-type regression vs tokenization output change; related area, not same bug." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD argument handling vs duplicated optimizer step under deepspeed; different failure modes." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Generation cache regression vs a testing-utils bug; unrelated." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "Precision warning during compile vs dtype not propagating on load; same theme, different code paths." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Missing image-patch helper vs MLA layernorm epsilon precision issue; unrelated." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "Save/sharding question vs Qwen3.5 MoE from_pretrained failure; not the same bug." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Missing projections in one model family vs a different Qwen3.5 MoE loading error." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Removed tie-embeddings API vs unexpected position-id key warnings; different regressions." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "torch.split return behavior vs RoPE implementation; too different." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "AutoImageProcessor import failure vs PretrainedConfig type-checking regression; different API breaks." + }, + { + "left": "issue:43012", + "right": "issue:43408", + "accept": false, + "reason": "bfloat16 precision warning vs SAM3 model-type mismatch warning; unrelated." + }, + { + "left": "issue:42222", + "right": "issue:45459", + "accept": false, + "reason": "VitPose model breakage vs tokenizer error masking without protobuf; no shared bug." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both involve flash attention, but different models and symptoms; not mergeable as one fix." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "Config type checking regression vs TensorBoard-only backend issue; unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch vs qwen3next missing projections; different concrete problems." + }, + { + "left": "issue:43335", + "right": "issue:45071", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction bug vs generic config type-checking regression." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Different Qwen3-VL model-loading failures; no evidence of the same root cause." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "Unsafe sys.modules access vs testing-utils bug; unrelated." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Embedding-tie deprecation vs gguf_file tokenizer/model loading breakage; different issues." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "AutoProcessor/torchvision crash vs testing-utils bug; unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer mapping None vs a specific model load failure; too little evidence of the same root cause." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "Video AutoProcessor crash vs max_shard_size question; unrelated." + }, + { + "left": "issue:43408", + "right": "issue:44368", + "accept": false, + "reason": "Model-type warning for sam3_tracker vs embedding-tie warning in Qwen3.5 fine-tuning; different configs." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Image patch helper missing vs torch.split return values; unrelated." + }, + { + "left": "issue:43976", + "right": "issue:45071", + "accept": false, + "reason": "Python version support regression vs config type-checking regression; different breakages." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Torchvision-dependent processor crash vs missing ReasoningEffort import; unrelated." + }, + { + "left": "issue:44829", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention training degeneration vs category cardinality limit; unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Qwen3-MoE sliding-window bug vs testing-utils bug; no overlap." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Different import errors from different modules; not the same failure." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "LayerNorm epsilon precision issue vs RoPE implementation discussion; distinct bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:36683", + "issue:38617", + "issue:42490", + "issue:42757", + "issue:42886", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43097", + "issue:43208", + "issue:43352", + "issue:43381", + "issue:43502", + "issue:43526", + "issue:43550", + "issue:43582", + "issue:43618", + "issue:43761", + "issue:43824", + "issue:43825", + "issue:43873", + "issue:43901", + "issue:43906", + "issue:43976", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44062", + "issue:44117", + "issue:44222", + "issue:44315", + "issue:44351", + "issue:44368", + "issue:44380", + "issue:44410", + "issue:44483", + "issue:44484", + "issue:44514", + "issue:44655", + "issue:44683", + "issue:44704", + "issue:45103", + "issue:45375", + "issue:45478", + "issue:45479", + "issue:45588", + "issue:45753" + ], + "soft_pairs": [ + "issue:43064|issue:44315", + "issue:38617|issue:44351", + "issue:44410|issue:45375", + "issue:44062|issue:44117", + "issue:43097|issue:45588", + "issue:44704|issue:45103", + "issue:44038|issue:44410", + "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:43550", + "issue:43352|issue:44380", + "issue:43012|issue:44368", + "issue:42886|issue:43381", + "issue:42907|issue:45753", + "issue:42907|issue:44655", + "issue:44380|issue:45479", + "issue:43352|issue:43976", + "issue:43064|issue:44380", + "issue:43208|issue:43526", + "issue:44483|issue:44514", + "issue:36683|issue:43582", + "issue:44222|issue:45478", + "issue:42490|issue:43012", + "issue:43618|issue:43761", + "issue:43012|issue:44683", + "issue:42757|issue:43824", + "issue:43618|issue:44380", + "issue:38617|issue:43502" + ], + "prepared_review_unit_hash": "aa33f3466cdd8039ab38eb536d8dd3e1f429f0101b9aa9f0dad44c6c2ff10797", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38058, + "estimated_input_tokens": 9515, + "estimated_eval_tokens": 19286 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose similarity cluster of mostly unrelated Transformers issues: import errors, model-specific regressions, tokenizer/pipeline bugs, and training/runtime edge cases. I do not see any soft pair that is clearly the same concrete bug/change, so all soft edges should be rejected.", + "confidence": 0.91, + "canonical_issue_reason": "No single canonical issue fits the whole cluster; the items span unrelated bugs and regressions rather than one shared defect.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the most active open item in the set and has broad user impact around quantized/offloaded loading, so it is the best representative anchor among these heterogeneous issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Different problems: distributed Trainer/FSDP2 optimizer-state corruption vs Liger Kernel application with model_init." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different missing symbols and different code paths." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Different model configs/fields: missing qwen3next projections vs missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizer-related, but one is AddedToken argument duplication and the other is TOKENIZER_MAPPING_NAMES returning None." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Unrelated regressions: removed tie_embeddings_and_encoder_decoder vs flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "Different utilities and failures: cached_file kwargs forwarding vs auto_docstring annotation handling." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Different Qwen model families and different failures; not the same underlying bug." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "One is offline/local_files_only network leakage; the other is a missing embed_tokens.weight on load." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Different areas: AutoProcessor/video-model dependency handling vs tokenizer mapping None handling." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Different issues: API removal warning vs quantization/offloading behavior." + }, + { + "left": "issue:43824", + "right": "issue:43986", + "accept": false, + "reason": "Both touch VL/processor loading, but one is an import failure and the other is a missing torchvision dependency crash." + }, + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Wrong tokenizer decoder type in v5 is unrelated to the isolated reproduction issue." + }, + { + "left": "issue:43097", + "right": "issue:43901", + "accept": false, + "reason": "API removal vs documentation/behavior mismatch for return_all_scores; not the same defect." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Different pipeline failures: misleading task support message vs inability to save Pipeline objects." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing model weight loading vs a max_shard_size question; no shared underlying bug." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "Different model/runtime bugs: xLSTM training blockers vs Bamba compile+SDPA failure." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A compile warning on bfloat16 models is not the same as a compile failure on Bamba with SDPA." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Different attention-related bugs in different models: unsupported Flash Attention 2.0 vs GPT2 attention scaling being ignored." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "Both mention warnings, but they concern unrelated behaviors and models." + }, + { + "left": "issue:42886", + "right": "issue:43381", + "accept": false, + "reason": "Offline cache loading failure vs gradient checkpointing in eval mode are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Saving dequantized models and an infinite-loop inference bug are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both mention saving, but one is dequantized model saving and the other is Pipeline.save_pretrained() support." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention-scaling backend bug vs degenerate zero-loss classification issue; no shared code-path." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Different regressions: Flash Attention support for a model vs Python version compatibility." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Distributed training optimizer-state corruption is unrelated to GPT2 attention scaling under SDPA/FlashAttention." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "Different model bugs: xLSTM training issues vs BeitImageProcessorFast label reduction behavior." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "API request handling in chat/completions is unrelated to Qwen2_5_VLProcessor batching/padding crashes." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size and an Apple Silicon allocator TypeError are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but different directions and failures: FP8 save_pretrained vs from_pretrained error on a newer version." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Local path trailing-slash import warning is unrelated to the bfloat16 compile precision warning." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP regressions, but one is missing attentions assignment and the other is hidden_states not returned; not the same concrete bug." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "Different compilation problems: a warning about precision vs a flex_attention failure on torch>=2.9." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Importing is_offline_mode from huggingface_hub is unrelated to a missing Qwen2_5_VLForConditionalGeneration import." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "Different model/code paths: CLIP output fields vs GPT2 attention scaling under backend attention implementations." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "Distinct regressions: a missing configuration_utils symbol vs network calls despite local_files_only=True." + } + ] + }, + "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-10000-400", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:41669", + "issue:42757", + "issue:42890", + "issue:42994", + "issue:43010", + "issue:43064", + "issue:43066", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43525", + "issue:43611", + "issue:43618", + "issue:43756", + "issue:43867", + "issue:43901", + "issue:43931", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44077", + "issue:44188", + "issue:44230", + "issue:44242", + "issue:44246", + "issue:44263", + "issue:44355", + "issue:44380", + "issue:44483", + "issue:44485", + "issue:44492", + "issue:44655", + "issue:44855", + "issue:44938", + "issue:45071", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45310", + "issue:45375", + "issue:45381", + "issue:45440", + "issue:45561", + "issue:45593", + "issue:45753" + ], + "soft_pairs": [ + "issue:44483|issue:45290", + "issue:44038|issue:44230", + "issue:43756|issue:44485", + "issue:44483|issue:45381", + "issue:43299|issue:43867", + "issue:43299|issue:45753", + "issue:43901|issue:44655", + "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:43066|issue:45381", + "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", + "issue:44188|issue:44380", + "issue:43064|issue:44492", + "issue:43352|issue:43931", + "issue:43010|issue:45278", + "issue:43992|issue:45310", + "issue:43867|issue:45375", + "issue:41669|issue:44246", + "issue:44355|issue:44855", + "issue:36683|issue:43064", + "issue:42890|issue:45561", + "issue:44938|issue:45071" + ], + "prepared_review_unit_hash": "890124d5fae0451df18f893dd9409c3eaf8da290a044502f5a168d2b43132670", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37171, + "estimated_input_tokens": 9293, + "estimated_eval_tokens": 18842 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a mixed bag of Transformers v5 regressions and model-specific bugs, not a single duplicate set. The shared wording mostly reflects broad themes like loading/import/config breakage, but the concrete failures differ too much for PR-style duplicate merging.", + "confidence": 0.96, + "canonical_issue_reason": "issue:45278 is the broadest umbrella here: it explicitly captures widespread import breakage after upgrading Transformers, which fits the cluster\u2019s overall regression pattern better than the more narrow model-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best global representative because it is the most general and user-facing summary of the upgrade regression theme in this cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Different failures: chat/completions request acceptance vs apply_chat_template crashing on assistant tool-call messages." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "One is a Qwen3-VL-Moe loading bug; the other is fp8 inference support for Qwen3-VL/Qwen3.5 MoE, so the code paths and fixes differ." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Both mention RoPE, but one is a Smollm3 layer-selection bug and the other is GLM-5 RoPE implementation discussion; not the same defect." + }, + { + "left": "issue:44483", + "right": "issue:45381", + "accept": false, + "reason": "Chat/completions request validation is unrelated to Qwen2.5-VL video vision_position_ids handling." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Qwen3-VL-Moe loading regression and a generic sorted-state_dict loading error are different model-load problems." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Model loading breakage and an infinite-loop-on-specific-image report are unrelated symptoms and likely different fixes." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Docs mismatch for return_all_scores is not the same as the pipeline save_pretrained serialization bug." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Gemma3 missing vocab_size and Sam3VisionOutput missing fpn_position_embeddings are different model attribute regressions." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure and pipeline object save_pretrained failure affect different save paths." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Nemotron FlashAttention2 support gating and GLM-5 GenerationConfig validation are unrelated." + }, + { + "left": "issue:44077", + "right": "issue:44380", + "accept": false, + "reason": "One is a patchtsmixer post_init policy change; the other is GPT2 attention scaling being ignored under SDPA/FlashAttention." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Generic state_dict ordering load error and Qwen3.5 save_pretrained regression are different load/save bugs." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Importing GenerationMixin from transformers.generation and importing is_offline_mode from huggingface_hub are distinct import regressions." + }, + { + "left": "issue:43066", + "right": "issue:45381", + "accept": false, + "reason": "Tokenizer decoder type in v5 and Qwen2.5-VL video position_ids are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Gemma3 config attribute missing and patchtsmixer post_init restriction have no shared code path." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Base_model_prefix loading regression is not the same as pipeline serialization with save_pretrained." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Python version compatibility regression and Python 3.14 load failure are too broad to treat as the same bug." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Gemma3 missing vocab_size and Llama4 missing pad_token_id are separate config-attribute regressions." + }, + { + "left": "issue:43352", + "right": "issue:45593", + "accept": false, + "reason": "FlashAttention2 support for Nemotron and D-FINE auxiliary-loss behavior are unrelated issues." + }, + { + "left": "issue:44263", + "right": "issue:45440", + "accept": false, + "reason": "torch.split return handling in GlmMoeDsaIndexer is not the same as DeepseekV3MoE diverging from remote implementation." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "Missing CLIPOutput attentions and Qwen3.5 save_pretrained corruption are different regressions." + }, + { + "left": "issue:36683", + "right": "issue:44242", + "accept": false, + "reason": "Gemma3 vocab_size missing is unrelated to router-logit load-balancing loss behavior." + }, + { + "left": "issue:44188", + "right": "issue:44380", + "accept": false, + "reason": "Torch.compile attention-kernel divergence and GPT2 attention scaling ignored in SDPA/FlashAttention are different backend bugs." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "Distributed optimizer-state corruption and a cache-strategy typo are not the same problem." + }, + { + "left": "issue:43352", + "right": "issue:43931", + "accept": false, + "reason": "FlashAttention2 unsupported for Nemotron and Qwen3-VL weight-shape mismatch are unrelated model-loading/runtime failures." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "no_grad decoration on cache updates is unrelated to broad import errors after upgrading to v5.5.0." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "UMT5 missing embed_tokens.weight and Qwen3.5 MoE from_pretrained errors are different model-load regressions." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "Generic sorted-state_dict loading error and missing deepstack_visual_indexes field are distinct issues." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "44246 is a vague slow-import report, but 41669 is a specific import-* removal/performance fix; insufficient evidence they are the same concrete change." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "SAM HQ set_seed flakiness is unrelated to xdist races on captured_info.txt." + }, + { + "left": "issue:44938", + "right": "issue:45071", + "accept": false, + "reason": "Python 3.14 load failure and PretrainedConfig type-checking breakage are different release regressions, not one underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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-10000-400", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:36296", + "issue:42490", + "issue:42548", + "issue:42831", + "issue:42890", + "issue:43010", + "issue:43065", + "issue:43232", + "issue:43299", + "issue:43502", + "issue:43575", + "issue:43618", + "issue:43723", + "issue:43867", + "issue:43874", + "issue:43976", + "issue:43992", + "issue:44062", + "issue:44075", + "issue:44077", + "issue:44112", + "issue:44222", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44484", + "issue:44485", + "issue:44556", + "issue:44749", + "issue:44857", + "issue:44938", + "issue:44964", + "issue:44977", + "issue:45125", + "issue:45341", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45440", + "issue:45478", + "issue:45600", + "issue:45850" + ], + "soft_pairs": [ + "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", + "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:43010|issue:43232", + "issue:30990|issue:31795", + "issue:45125|issue:45375", + "issue:30333|issue:33453", + "issue:43299|issue:44556", + "issue:43502|issue:44222", + "issue:42831|issue:43874" + ], + "prepared_review_unit_hash": "53f14646eae41d8dcb897d8b34f53dca965f70842fb83ee84170c1736180031c", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37013, + "estimated_input_tokens": 9254, + "estimated_eval_tokens": 18764 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly unrelated issue reports, with a few thematically similar loading/regression bugs but no true duplicate-level matches among the soft pairs.", + "confidence": 0.66, + "canonical_issue_reason": "issue:43502 is the clearest, most actionable bug report in the set: it describes a concrete, user-visible regression with an obvious repro and fix target.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43502 is the best issue anchor because it is specific, reproducible, and directly actionable rather than broad or model-specific.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Different failures: state_dict ordering during model loading vs incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "One is a specific no_grad decoration request; the other is a broad complaint that many models break." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Both are loading problems, but they target different models and different root causes." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Unrelated: cache update decoration vs a Python-version compatibility failure." + }, + { + "left": "issue:44938", + "right": "issue:45381", + "accept": false, + "reason": "Different areas: Python 3.14 import/load failure vs Qwen2.5-VL video position-id behavior." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "Different model/API issues: CLIPOutput attentions vs patchtsmixer post_init policy." + }, + { + "left": "issue:42490", + "right": "issue:44075", + "accept": false, + "reason": "Local model path warning and SGD optimizer arguments are unrelated bugs." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Both mention tensor parallelism, but one is a generic training bug and the other is a specific OOM on loading a Qwen model." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "Different code paths: DSA indexer activation issue vs DeepseekV3MoE implementation divergence." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "ImportError for missing PyTorch and tokenizer loading regression are not the same defect." + }, + { + "left": "issue:43867", + "right": "issue:44484", + "accept": false, + "reason": "Model load failure from sorted state_dict is unrelated to the save_pretrained shard-size question." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "CLIPOutput attentions assignment and AMP/CUDA loss crash are different bugs." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "Different subsystems and symptoms: CLIPOutput field regression vs Qwen3.5 flash-attention generation bug." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "A flaky test due to missing set_seed is unrelated to a model-loading failure." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are CI/test-related, but the failure modes and affected components differ." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test seeding issue and auto_mappings stale config references are separate problems." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "Local_files_only network leakage and AddedToken keyword collision are unrelated." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both involve version changes and loading, but one is tokenizer loading and the other is checkpoint reload after upgrade." + }, + { + "left": "issue:42890", + "right": "issue:45341", + "accept": false, + "reason": "A flaky model integration test is not the same as a generic testing_utils bug." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained behavior and missing tensor-parallel plan are different feature gaps." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Int4 memory regression and checkpoint reload incompatibility are different issues." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different export fixes." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "No_grad on update methods and a dummy Conv2d in Sam3PixelDecoder are unrelated." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "local_files_only network requests and GLM-5 RoPE implementation are unrelated defects." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled-Python-file errors and post-upgrade inference slowdown are different problems." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5 MoE-related, but one is missing tensor-parallel metadata and the other is a generic from_pretrained failure." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Generation state update timing and no_grad decoration are different code paths." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Stuck model loading and documentation confusion are not the same bug." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Qwen3.5 tensor-parallel metadata and missing vision config field are separate configuration bugs." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLFlow reporting on failed jobs and tokenizer loading regression are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Both mention upgrade-related loading regressions, but they affect different models and failure modes." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "Network access despite local_files_only and FP8 save_pretrained behavior are not duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:30333", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:41669", + "issue:42371", + "issue:42548", + "issue:42757", + "issue:43502", + "issue:43525", + "issue:43526", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43824", + "issue:43827", + "issue:43901", + "issue:44038", + "issue:44222", + "issue:44230", + "issue:44261", + "issue:44273", + "issue:44315", + "issue:44355", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44655", + "issue:44704", + "issue:44908", + "issue:44938", + "issue:44961", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45125", + "issue:45230", + "issue:45362", + "issue:45468", + "issue:45542", + "issue:45753", + "issue:45850" + ], + "soft_pairs": [ + "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:43824", + "issue:43576|issue:45071", + "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", + "issue:44273|issue:45542", + "issue:44273|issue:45003", + "issue:44393|issue:45753", + "issue:44393|issue:45071", + "issue:44038|issue:45125", + "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" + ], + "prepared_review_unit_hash": "f45d1b0f4a8716cd6529002bf46596050cd19ee182ccedd680a8836fc92f822d", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 39, + "serialized_chars": 38478, + "estimated_input_tokens": 9620, + "estimated_eval_tokens": 19496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a collection of unrelated Transformers v5-era regressions and import/loading complaints. Issue 45850 is the broadest umbrella, but most pairs are not true duplicates; they only share a generic theme like 'v5 broke something.'", + "confidence": 0.81, + "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set and best matches the overall 'v5 breaks many models' regression theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45850 is the best single anchor for the cluster, though it is too vague to serve as a precise duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Different subsystems and symptoms: model_init/Liger Kernel vs DSA indexer/ReLU." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "Unrelated issues: offline/local_files_only behavior vs an empty 'racoon' report." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "One is about compiled Python file loading; the other is PretrainedConfig type checking." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLFlow reporting on failed training jobs is unrelated to tensor-parallel training behavior." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled Python file errors and pipeline save_pretrained failures are different bugs." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs mentioning pipeline() after v5 removals is unrelated to the 'racoon' issue." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "Both mention FP8/MoE, but one is save_pretrained behavior and the other is inference support." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Both are loading regressions, but they point to different code paths and a specific Llama Vision case." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "ImportError from huggingface_hub is unrelated to the v5 env command breakage." + }, + { + "left": "issue:33453", + "right": "issue:36296", + "accept": false, + "reason": "Tokenizer loading regression is unrelated to tensor-parallel training." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Broken env command and missing Qwen2.5-VL export are separate failures." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Env command breakage is unrelated to PretrainedConfig type-checking regressions." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Documentation confusion in forward args is unrelated to an ImportError for layer_type_validation." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "A placeholder issue and Gemma audio positional encoding bug are not duplicates." + }, + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different export gaps." + }, + { + "left": "issue:38617", + "right": "issue:43576", + "accept": false, + "reason": "ImportError for layer_type_validation and env command breakage are different problems." + }, + { + "left": "issue:43576", + "right": "issue:44938", + "accept": false, + "reason": "Broken env command is not the same as Python 3.14 load failures." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Chunked_prefill cache issue and a cache-strategy typo are not the same bug." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs handling is unrelated to the placeholder 'racoon' issue." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "RoPE implementation discussion is unrelated to the placeholder issue." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training bug is unrelated to documentation about return_all_scores." + }, + { + "left": "issue:41628", + "right": "issue:44355", + "accept": false, + "reason": "AutoImageProcessor import error and compiled-file loading errors are different regressions." + }, + { + "left": "issue:36296", + "right": "issue:43526", + "accept": false, + "reason": "Tensor parallel training and BeitImageProcessorFast label reduction are unrelated." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Qwen3.5 chat crash is a specific model/runtime bug; 45850 is only a broad umbrella." + }, + { + "left": "issue:44261", + "right": "issue:44360", + "accept": false, + "reason": "MLA rms_norm_eps precision issue is unrelated to the DSA/ReLU report." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Missing processor.save_pretrained files is unrelated to the placeholder bug report." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading problems and tensorboard/TensorFlow backend errors are different failures." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading behavior and unsafe sys.modules access are related only broadly, not as the same concrete bug." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Different Qwen model/version issues: 2D box output vs infinite loop on one image." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "Qwen3-VL output bugs are unrelated to PretrainedConfig type-checking." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Generic Qwen3-VL-MoE bug report and missing _tp_plan for tensor parallelism are distinct." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Env command breakage is unrelated to checkpoint reload incompatibility after upgrade." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Missing pad_token_id in Llama4Config is too specific to be the same as the broad v5 breakage umbrella." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "Cache strategy typo is not the same as the broad 'v5 breaks too many models' complaint." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLmConfig pad_token_idx regression is a separate model-specific issue." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 behavior settings and cache strategy typo are unrelated." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "Missing kwargs in AutoProcessor.from_pretrained is a different code path from scheduler kwargs handling." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 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-10000-400", + "nodes": [ + "issue:31795", + "issue:33453", + "issue:34689", + "issue:37428", + "issue:41669", + "issue:42371", + "issue:42890", + "issue:42913", + "issue:43066", + "issue:43352", + "issue:43408", + "issue:43502", + "issue:43519", + "issue:43576", + "issue:43606", + "issue:43704", + "issue:43827", + "issue:43874", + "issue:43976", + "issue:44077", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44623", + "issue:44655", + "issue:44829", + "issue:44861", + "issue:44908", + "issue:44945", + "issue:44998", + "issue:45071", + "issue:45092", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45850" + ], + "soft_pairs": [ + "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:43576|issue:44908", + "issue:44297|issue:45230", + "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:44315|issue:44829", + "issue:41669|issue:44908", + "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", + "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" + ], + "prepared_review_unit_hash": "d2b11dbba0b97c79375e58757a758ab2c19c1709d569f65a3a1961b5a71158dd", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 39635, + "estimated_input_tokens": 9909, + "estimated_eval_tokens": 20074 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mix of Transformers v5 regression reports across tokenizers, imports, loading, and model-specific failures. issue:45278 remains the best umbrella issue, but the soft edges should stay conservative: none of these pairs are clearly the same concrete bug from the provided titles alone.", + "confidence": 0.58, + "canonical_issue_reason": "issue:45278 is still the strongest canonical issue: it is open and explicitly captures the broad multi-model breakage after the 4.57.0 -> 5.5.0 upgrade, making it a practical umbrella for this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best global issue target because it is open, broad enough to subsume related v5 breakage reports, and specific enough to remain useful as a tracker.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "SamHQ test flakiness from missing set_seed is unrelated to inverse_sqrt scheduler kwargs handling." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Both are broad v5 breakage complaints, but they do not describe the same concrete failure mode." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Import slowness and lazy-loading failure are related themes, but not clearly the same bug." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "Model-type warning for sam3_tracker is unrelated to Liger Kernel application in model_init." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version compatibility and Qwen3-VL bbox output errors are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2 support error is unrelated to Qwen3-VL bounding-box hallucination/output bugs." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "Broken transformers env command is unrelated to scheduler kwargs being ignored." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Specific tokenizer_class mismatch is too different from a generic bug report to treat as the same issue." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Position_ids buffer access error is not the same as an unspecified generic bug report." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs bug and PretrainedConfig type-checking regression are different code paths." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-* performance regression is unrelated to the HybridCache import error." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Import-* performance regression is unrelated to a testing_utils bug." + }, + { + "left": "issue:44492", + "right": "issue:44945", + "accept": false, + "reason": "Cache-strategy typo and pipeline-parallelism output corruption are unrelated." + }, + { + "left": "issue:43874", + "right": "issue:45850", + "accept": false, + "reason": "Missing get_number_of_image_patches is a concrete GLM46V bug, not the same as generic model breakage." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Qwen3-VL bbox issues and Qwen3.5 MoE from_pretrained errors are different failures." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "patchtsmixer post_init policy and tied-weight-key AttributeError are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs still referencing pipeline() is unrelated to inverse_sqrt scheduler kwargs." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Confusing forward docs and tokenizer loading regression are different issues." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel not applied with model_init is unrelated to flash_attention_3 degenerate training." + }, + { + "left": "issue:41669", + "right": "issue:44908", + "accept": false, + "reason": "Import-* performance regression is unrelated to a scheduler keyword bug." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Broken env CLI and lazy-loading failure are only loosely related." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Pipeline save_pretrained failure is unrelated to _tied_weight_keys list handling." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "DSA indexer ReLU issue and Gemma-4 audio positional encoding bug are different model-specific problems." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "A generic bug report is too vague to confirm it is the same as the Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading and inverse_sqrt scheduler kwargs are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different model bugs." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Timestamp calculation in Qwen3VL is unrelated to scheduler kwargs handling." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import-* performance regression is unrelated to a VRAM leak in dataloader threads." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Model.forward documentation confusion is unrelated to a Llama 3.2 model-loading regression." + }, + { + "left": "issue:31795", + "right": "issue:37428", + "accept": false, + "reason": "Forward-doc confusion is unrelated to the flash-attention import error." + }, + { + "left": "issue:44315", + "right": "issue:45092", + "accept": false, + "reason": "Both involve model initialization, but they describe different concrete incompatibilities." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Different models and different bugs: GLM-5 RoPE versus Gemma-4 audio positional encoding." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak in multi-threaded dataloading is unrelated to GLM-5 RoPE issues." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 configuration guidance is unrelated to a VRAM leak." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "local_files_only network requests and Qwen3VL timestamp calculation are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Timestamp calculation bug and processor.save_pretrained missing files are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Timestamp calculation bug and VRAM leak are unrelated." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "torch.split return-values issue and Gemma-4 audio positional encoding bug are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch and Gemma-4 audio encoding issue are unrelated." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "LayerNorm epsilon precision issue and Gemma-4 audio encoding issue are unrelated." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "The titles point to two different tokenizer-v5 regressions: one is a broad behavior difference, the other is a specific wrong decoder-type bug. That is not enough to confirm the same underlying issue." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs bug is unrelated to the placeholder/unemployment issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster summary is mostly grounded, and the 45278 umbrella choice is plausible, but I can\u2019t accept the output because one soft-edge verdict is too strong: 42913 \u2194 43066 is marked true even though the packet only gives titles showing two different tokenizer-v5 regressions (general behavior difference vs wrong decoder type), which is not enough to confirm the same bug. That makes the overall judgment insufficiently conservative." + }, + "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": 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-10000-400", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:36296", + "issue:38617", + "issue:43298", + "issue:43525", + "issue:43572", + "issue:43618", + "issue:44230", + "issue:44492", + "issue:44623", + "issue:44861", + "issue:44998", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45850" + ], + "soft_pairs": [ + "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": "111f8039cce9a3727b6939c68c609b23c223a5569a22976c763bf66425840a97", + "component_budget": { + "node_count": 400, + "item_count": 400, + "soft_pair_count": 1499, + "serialized_chars": 772968, + "estimated_input_tokens": 193242, + "estimated_eval_tokens": 386740 + }, + "budget": { + "node_count": 19, + "item_count": 19, + "soft_pair_count": 15, + "serialized_chars": 15089, + "estimated_input_tokens": 3773, + "estimated_eval_tokens": 7802 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: it mixes unrelated regressions, a docs typo, a model-config attribute bug, and several loading/runtime issues. None of the soft pairs look like the same underlying bug, so I would not merge any of them as duplicates.", + "confidence": 0.19, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45310 is the most concrete and actionable regression report in the set (a specific from_pretrained failure on Qwen3.5 MoE), but the surrounding issues are too diverse for a true canonical duplicate.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Different failures: tied-weight key handling vs chat template parsing with tool calls. No shared code-path problem." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "A cache-strategy typo is unrelated to an unemployment report." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "Generic broad complaint about many models breaking is not the same as an unemployment issue." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIP attentions assignment regression is unrelated to the unemployment report." + }, + { + "left": "issue:44998", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated subjects: unemployment vs a testing_utils bug." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Llama4Config pad_token_id absence and tied-weight key crash are different model/config issues." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "StableLmConfig pad_token_idx regression is not the same bug as _get_tied_weight_keys crashing." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow from_pretrained checkpoint loading and an ImportError for layer_type_validation are unrelated." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 Qwen/VL support request is different from processor.save_pretrained missing files." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLFlow job-status reporting and checkpoint loading performance are unrelated." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Checkpoint loading slowness is not the same as a tensor-parallel training bug." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Loading speed regression and documentation confusion are different classes of issue." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen/from_pretrained-like contexts, but one is FP8 support and the other is a specific loading error; not the same concrete bug." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Tokenizer loading regression is distinct from slow checkpoint loading." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching audio-model support and FP8 Qwen/VL support are different feature gaps." + } + ] + }, + "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": "f4e96e0d13b14c57a4991de17e4204714c51358c4c93d0cefc666cfd89a07c13", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster only has loose topical similarity around model/token handling. Issue 13244 is about tokenization mismatch, 15980 is about an auth/download error message, and 16618 is a model loading failure. They do not look like duplicates of the same underlying bug.", + "confidence": 0.86, + "canonical_issue_reason": "issue:16618 is the most concrete user-facing runtime failure in the cluster and the broadest model-loading problem; the others are narrower and about different code paths.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:16618 is the best anchor for the cluster because it is the clearest concrete bug report, but it still does not subsume the other two issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tokenization output differing from TensorFlow code is a different bug class from failing to load a model; the symptoms and likely code paths do not match." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Bad error messaging for unauthenticated private-model download is not the same as a generic 'can't load bert-base-uncased' failure; related area, but not the same underlying defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "3b7ac85596561b41adac32f95e86a3080e054d8dcc787e3bde3d11856dfe6f62", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate community-contribution feature requests: one for Flash Attention 2 architecture support and one for GGUF architecture support. They share a broad theme but not the same underlying bug/change.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the requests target different features and different code paths.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a duplicate of the other, so there is no single canonical issue for the pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different requested capabilities (Flash Attention 2 vs GGUF support); same broad subsystem, but not the same underlying change or bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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-27088-4", + "nodes": [ + "issue:27088", + "issue:43834", + "issue:44365", + "issue:44370" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:44365|issue:44370" + ], + "prepared_review_unit_hash": "1dff98e9c7c67da2e0c92448858e5db7da25cc236f3d9605926697bc785fafb8", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4452, + "estimated_input_tokens": 1113, + "estimated_eval_tokens": 2482 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4452, + "estimated_input_tokens": 1113, + "estimated_eval_tokens": 2482 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These look like translation-request issues, but only one is concrete (Turkish). The others are generic template-style placeholders and may represent different language requests, so I would not collapse them as duplicates.", + "confidence": 0.74, + "canonical_issue_reason": "Issue 27088 is the most concrete and specific: it names Turkish explicitly and looks like a real translation request rather than a template placeholder.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 27088 is the strongest representative because it is specific, actionable, and not just a generic language-template title.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Turkish-specific request vs a generic language-template issue; the similarity appears to come from the shared template, not the same underlying request." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Turkish-specific request vs a generic language-template issue; no evidence they target the same language or same translation task." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Turkish-specific request vs a generic language-template issue; similar wording alone is not enough to treat them as the same issue." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "Both appear to be template-style translation requests, but there is no concrete evidence they are the same language or the same underlying task." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "Identical placeholder-style titles are not enough to merge; they may correspond to different language-specific requests." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "Both are generic translation-request templates, but no explicit language target or shared concrete bug/change is shown." + } + ] + }, + "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": "ea29828d0bdc27ecc969455f3ab60d3576f0860d8e95d5078852e49cbba68b75", + "component_budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 35, + "serialized_chars": 20712, + "estimated_input_tokens": 5178, + "estimated_eval_tokens": 10612 + }, + "budget": { + "node_count": 15, + "item_count": 15, + "soft_pair_count": 35, + "serialized_chars": 20712, + "estimated_input_tokens": 5178, + "estimated_eval_tokens": 10612 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a loose theme cluster of model-add/support requests, not true duplicates. The issues name different models or different modalities, so they should stay separate; at most, a couple are nearby subfamily requests.", + "confidence": 0.84, + "canonical_issue_reason": "If one issue must act as the anchor, issue 43837 is the best fit because it sits in the most connected subcluster and has the strongest overlap with a nearby related request (43671), but the set is still not duplicate enough to collapse into one issue.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43671 is the best representative of the cluster because it is a clear, specific model-support request with substantial discussion and belongs to the densest nearby subcluster; however, it still represents a distinct request from the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different model requests (EuroBert vs DEIMv2); same broad pattern, not the same bug/change." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "Different requested models and different modality area; not a duplicate." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "Both are model-add requests, but for unrelated models and code paths." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are distinct model integrations." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "Detection model request vs audio model request; not the same change." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "Different model families and integration work; no single concrete fix." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "Different requested models; same support-request theme only." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR vs Qwen3-TTS are unrelated model additions." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are separate model support requests." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "Different models with different implementation paths." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "NeoBERT and LFM2.5 Audio are unrelated requests." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "Different models; only the generic request template matches." + }, + { + "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 request vs Qwen3-ASR support; different artifacts." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "Different model families and likely different files/code paths." + }, + { + "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 architecture vs Voxtral TTS support; not the same change." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "DEIMv2 and Qwen3-TTS are unrelated model requests." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "ModernVBERT and Qwen3-ASR are different models." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT models vs Qwen3-TTS support; not a duplicate." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "ModernVBERT and BERT-like models with RoPE are different requests; too broad to merge." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "A generic BERT-like/RoPE request is not the same as Voxtral-4B-TTS support." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Same Qwen3 family, but TTS vs ASR are different products and code paths." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Both are speech-related, but they request different model support." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated model additions." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Different model families and different integration work." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-TTS and LFM2.5 Audio are distinct audio-model requests." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-ASR and LFM2.5 Audio are not the same model or fix." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR support and EUPE are unrelated model requests." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "Qwen3-ASR and the weight-sparse transformer are distinct requests." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "Audio model support vs SAM3-LiteText; different model integrations." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Weight-sparse transformer and EUPE are separate model requests." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Both are speech-related, but ASR and TTS support are different concrete changes." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral-4B-TTS are unrelated model additions." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-ASR and DeepSeek-OCR2 do not target the same model or code path." + } + ] + }, + "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": "6f142b0557b4ef80729795269016402f28030efc2717b63aa38ccea90564d478", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 6291, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 8, + "serialized_chars": 6291, + "estimated_input_tokens": 1573, + "estimated_eval_tokens": 3402 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues are only loosely text-similar and describe different bugs across unrelated areas (model norms, config typing, loss kwargs, a typo, tokenizer warnings). I do not see a true duplicate cluster here.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue: the items cover unrelated problem types and code paths, so none is a good duplicate anchor.", + "canonical_pr_reason": null, + "best_issue_reason": "No best issue for deduplication because the set is heterogeneous rather than one underlying bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Both mention normalization/config names, but one questions an architecture choice in Llama4 while the other is a type-hint issue in ModernBERTConfig; not the same bug." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Mask2Former ignore_value handling and loss kwargs plumbing are different parts of the training stack and different failure modes." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Unrelated topics: a model loss parameter not used versus tokenizer regex warning regression." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "A Mask2Former loss-parameter issue is unrelated to a typo in a quantization-operation attribute name." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Different models and different bugs: ignore_value usage versus an incorrect type hint." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Loss-kwargs propagation and ModernBERTConfig type annotation are separate issues with no shared code path." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "A config type hint bug does not match a tokenizer-regex warning regression." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "ModernBERTConfig typing and an attribute-name typo are distinct problems in different code areas." + } + ] + }, + "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": "f2f820abf20784a482a6536760011db0bb596b7a6121c03cc6d888fbfaf5586e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1478, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1478, + "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 is a vague feature request for \"Set Block Decoding,\" while the other is a specific request for \"DeepSeek v3.2 support.\" They appear to be unrelated product/runtime feature requests rather than the same underlying problem.", + "confidence": 0.92, + "canonical_issue_reason": "Issue #43037 is the better canonical anchor because it is more specific and actionable; #41084 is too vague to serve as a stable representative for the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "#43037 is the best standalone issue candidate due to its concrete scope and clearer target. #41084 is too underspecified to be the primary reference issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "The titles and available metadata point to different requests: block decoding behavior versus DeepSeek v3.2 model support. No shared files, targets, or issue links suggest the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 8, + "component_count": 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": "ea4e7f13aec445d58bcdb636eb9f774f112fcc06e0aa978d7f87c8862aa99c46", + "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 issues are about different trainer behaviors: one concerns `remove_unused_column` customizability, the other concerns FLOPS estimation for custom trainers. They do not appear to describe the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Different problem domains and code paths: one is about column pruning customization in Trainer, the other about FLOPS accounting for custom trainers. Shared theme is only broad Trainer customization, not a duplicate 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": "c42d55c30f6c06de9da09976bc911fd47df1406b9fad1a8084703601c9c1e033", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both issues are about extending Hugging Face auto-class support for Qwen-family models, but they request different changes: Qwen3-Omni registration in core auto classes vs. Qwen3.5 sequence-classification support. They are related feature requests, not the same underlying bug/change.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43684 is the broader auto-registration request and best represents the cluster\u2019s general theme, but it is still not the same as the Qwen3.5 sequence-classification request.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43684 is the better cluster representative because it targets core AutoModel/AutoModelForConditionalGeneration registration rather than a narrower task-specific head.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different model variants and different auto-class targets; related in theme, but not the same concrete issue or change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "46073e5cddb5c8499c487ae96083e3eca4e454d69aaeae59b362c4f3cfc03ee8", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4258, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4258, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "#43845 is the best canonical issue among the three near-identical \"huggingface \" reports because it was opened first. #43835 looks unrelated. The similarity between the huggingface issues is too shallow to confidently mark any pair as duplicates without body details or explicit links.", + "confidence": 0.78, + "canonical_issue_reason": "#43845 is the earliest of the three near-identical \"huggingface \" issues, so it is the best representative of that cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "#43845 is the strongest issue candidate because it is the first report in the placeholder-title cluster and is the most natural canonical representative.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles and no concrete shared bug context; the evidence is only weak text similarity." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Different titles and no shared details point to unrelated reports rather than the same issue." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "\"Fett\" does not plausibly describe the same underlying problem as the huggingface issues." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": false, + "reason": "Same placeholder title, but only soft similarity is available and deterministic acceptance is false, so this is not enough to confirm a duplicate." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "The titles are similar, but there are no body details or explicit links tying them to the same underlying bug." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "These look superficially alike, but the available evidence is too weak to treat them as duplicates with confidence." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The canonical issue choice is reasonably grounded: #43845 is the earliest of the three issues with the same title. However, the soft-edge verdicts are too strong for the packet: all pairings are supported only by soft_similarity, and the packet\u2019s deterministic_accept is false for every pair. With no issue-body details or explicit links, it is not conservative to mark the #43845/#43846/#43859 pairs as duplicates with high confidence. Rejecting the #43835 comparisons is fine, but the duplicate conclusions among the huggingface issues overstate the evidence." + }, + "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": "5240be50cddc1c18171a3108f04aac3f19fa353feaeeb470bc99c61a7ce9c13e", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are not clear duplicates of one bug/feature; they look like unrelated low-signal/spam issues with only superficial textual similarity. No soft pair should be merged.", + "confidence": 0.87, + "canonical_issue_reason": "issue:44418 is the least spam-like and most issue-shaped title in the set, so it is the best forced representative if one must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44418 is the most plausible standalone issue title here; the others are generic spam placeholders or unrelated text.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are just '' with no concrete shared bug or change; no evidence they describe the same underlying issue." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Both are spam placeholders and provide no shared problem description beyond the title text." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "One is spam and the other is a generic documentation suggestion; not the same underlying issue." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "'' and 'Check out \"Google LLC\"' are unrelated titles with no concrete bug overlap." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Documentation enhancement request does not match a spam placeholder issue; no shared problem statement." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Generic documentation request and spam placeholder are unrelated." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Same spam marker, but that alone is not enough to indicate the same underlying bug or change." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder and unrelated Google LLC title do not point to the same issue." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "Spam placeholder and unrelated Google LLC title do not describe the same bug or request." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "44f6144620f78e303387c8dab336fc1e19178192af2e148f87351790478c33c9", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1645, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1645, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are highly similar and may be duplicates: they share the exact title, have very close body lengths, and there are no distinguishing links, targets, or filenames. However, the packet does not prove they are exact duplicates, so the safer conclusion is likely-duplicate rather than confirmed duplicate. #45145 is the better canonical record because it was created first and has an inbound reference.", + "confidence": 0.83, + "canonical_issue_reason": "#45145 is the best canonical issue because it is earlier and has an inbound reference. The evidence supports strong similarity with #45151, but not a definitive exact-duplicate determination.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is the best representative issue for the cluster: it is the earlier report and has slightly more established history via the inbound reference, while #45151 does not add distinct scope.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "The pair looks highly similar, but the evidence is only soft similarity plus matching titles. With no explicit links, targets, or content proof, this is not enough to confidently accept them as the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The canonical choice is plausible, but the writeup overstates the evidence. The packet supports that the two issues have the same title and very similar metadata/body lengths, with no explicit links or targets, but it does not establish that they are exact duplicates or that the issue content is the same. The soft pair is also only a similarity signal and the deterministic decision is false, so the duplicate verdict and 0.99 confidence are too strong. A more conservative summary would say they appear highly similar / likely duplicates, with #45145 as the better canonical record because it is earlier and has an inbound reference." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": false, + "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": "65b21d096078d27c4cceb409f978dd81a76bcc6de62d377dfa17d102c8c99c21", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1668, + "estimated_input_tokens": 417, + "estimated_eval_tokens": 1090 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1668, + "estimated_input_tokens": 417, + "estimated_eval_tokens": 1090 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues both mention 4-bit inference, but they describe different failure modes and likely different code paths: one is a Gemma4/audio-module uint8 dtype error, the other is a BitsAndBytesConfig skip-module regression that removes the default lm_head exclusion. I would keep them separate and not merge the cluster.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45674 is the better cluster representative because it describes a concrete, broadly applicable BitsAndBytesConfig regression with a clear expected/actual behavior and an obvious configuration-level fix path.", + "canonical_pr_reason": null, + "best_issue_reason": "#45674 is the strongest issue to anchor on: it is specific, reproducible, and appears to describe the main actionable bug in the cluster. #45672 is a narrower model-specific dtype crash in audio modules and does not match the same root cause.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Both are 4-bit inference bugs, but the failure mechanisms differ: #45672 is a Gemma4/audio uint8 torch.finfo TypeError, while #45674 is a BitsAndBytesConfig skip-module bug causing lm_head exclusion to be cleared and an AssertionError. They do not look like 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": 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": "7d1384808e2e81500921827e3a7a187a03629af2c2f7ed4f132e07478de6c054", + "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": "Both items look like unrelated low-information issues/spam entries (personal names plus birth dates), not duplicate bug reports. There is no meaningful canonical issue or PR here.", + "confidence": 0.97, + "canonical_issue_reason": "None: neither issue contains a bug report, repro, or substantive discussion to serve as a canonical duplicate target.", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is suitable as a global best issue; both are empty placeholder-like entries with no actionable content.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "The titles are just different personal names and dates; there is no shared bug, feature, or code-path problem to indicate duplication." + } + ] + }, + "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": "110af76adcf80338b6b3aa8304f9685c59f559825f900f0c9c81faedbffdeb5a", + "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 issues are just separate name-and-date titles with no body, links, or shared context; there is no evidence they describe the same bug or request.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "No substantive overlap beyond both being bare personal-name titles; they do not appear to report the same underlying problem." + } + ] + }, + "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": "d44579cb77a0837216217ddc5acd11ab1afa90d63ab85500a28997e9dc6d6dee", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two exact duplicate issue pairs are present: 45870/45871 and 45872/45873. Issue 45869 appears to be a separate dataset/sample request and does not match the others.", + "confidence": 0.97, + "canonical_issue_reason": "45870 is the earliest of the exact duplicate pair 45870/45871 and has the cleanest representative title for that topic.", + "canonical_pr_reason": null, + "best_issue_reason": "45870 is the best single issue to keep for its duplicate pair because it is the earliest, clear, and exact match to 45871.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Different requests: a financial dataset bundle versus a DataForge prediction example." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Different requests: a financial dataset bundle versus a DataForge prediction example." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Exact duplicate titles and same described example request." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Different topics: DataForge prediction example versus DataForge plus scraping service." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Exact duplicate titles and same described DataForge scraping-service request." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The summary conservatively separates 45869 from the two duplicate-looking pairs, and the pair verdicts match the packet: 45870/45871 and 45872/45873 have identical titles and jaccard 1.0, while the cross-topic pairs are reasonably rejected." + }, + "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-39895-16", + "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:45899", + "pull_request:45919", + "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:45919", + "pull_request:43424|pull_request:45899", + "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:45899", + "pull_request:44827|pull_request:45899", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:45795", + "pull_request:44827|pull_request:45655", + "pull_request:45170|pull_request:45899", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:45213|pull_request:45899", + "pull_request:45795|pull_request:45899", + "pull_request:43823|pull_request:45795", + "pull_request:43823|pull_request:45899", + "pull_request:44827|pull_request:45922", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "a33177a03aea6443bb15300d6f1cb27231bb9e303e6daebbc911e61023395bd2", + "component_budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 29, + "serialized_chars": 46699, + "estimated_input_tokens": 11675, + "estimated_eval_tokens": 23606 + }, + "budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 29, + "serialized_chars": 46699, + "estimated_input_tokens": 11675, + "estimated_eval_tokens": 23606 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous: model additions, docs updates, refactors, and unrelated fixes. Only one true duplicate pair stands out: the two `fixed_cross_entropy` PRs that both implement the same kwargs support in `src/transformers/loss/loss_utils.py` for issue 43240.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43251 is the clearest canonical representative of the only real duplicate pair: it targets issue 43240 and makes the concrete `fixed_cross_entropy` kwargs change in `loss_utils.py`.", + "best_issue_reason": null, + "best_pr_reason": "PR 43251 is the strongest single PR to keep as representative because it directly fixes the linked loss issue with the exact same code-path change as 43254.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "Unrelated changes: VideoPrism model/docs addition vs ExecuTorch dynamic-shape export test." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Different artifacts: Mistral tokenizer method additions vs adding a new MobileLLM model." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Tokenizer API additions vs ExecuTorch export test; no shared bug or change." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor robustness fix vs new model implementation; not the same underlying change." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Loss utility kwargs fix vs MobileLLM model add; unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "Both touch training-related code, but cross-entropy kwargs support and Mistral4 test fixes are different changes." + }, + { + "left": "pull_request:43424", + "right": "pull_request:45919", + "accept": false, + "reason": "ExecuTorch export test and Sapiens2 model addition are unrelated." + }, + { + "left": "pull_request:43424", + "right": "pull_request:45899", + "accept": false, + "reason": "ExecuTorch exportability test vs docs-only continuous batching update; different code paths." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Dynamic-shape export test vs Mistral4 test/model work; no shared fix." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "ExecuTorch export testing and new model scaffolding are unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "New MobileLLM model vs config-resolution ordering fix; not the same bug." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45922", + "accept": false, + "reason": "New model addition vs lru-cache memory leak fix in vision models; different changes." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "New model addition vs Falcon output-collection refactor; unrelated." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor vs model-creation skill docs; not the same artifact or fix." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon interface refactor and Mistral4 test fix are different code-path changes." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45899", + "accept": false, + "reason": "Falcon model refactor vs docs-only decode fast-path note; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45899", + "accept": false, + "reason": "Mistral4 test/model work vs docs-only continuous batching update; no common bug." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 test fix vs CLIP-family layernorm typo mapping; different changes." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45795", + "accept": false, + "reason": "Mistral4 test/model changes vs docs for audio/video processors; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 work vs config resolution order fix; different underlying issue." + }, + { + "left": "pull_request:45170", + "right": "pull_request:45899", + "accept": false, + "reason": "Layernorm typo rename vs docs-only batching fast-path description; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4 tests/model code vs model-creation skill docs; not the same change." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "New MobileLLM model vs model-creation skill docs; no shared code-path problem." + }, + { + "left": "pull_request:45213", + "right": "pull_request:45899", + "accept": false, + "reason": "Skill docs for model creation vs continuous batching docs update; unrelated." + }, + { + "left": "pull_request:45795", + "right": "pull_request:45899", + "accept": false, + "reason": "Docs for audio/video processors vs docs for decode fast path; different documentation topics." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45795", + "accept": false, + "reason": "New MobileLLM model vs docs for processor components; unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45899", + "accept": false, + "reason": "New MobileLLM model vs decode fast-path docs; no common underlying change." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45922", + "accept": false, + "reason": "Mistral4 test/model work vs vision-model cache-leak fix; unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Same file, same function, same issue 43240, and both add the same `cross_entropy` kwargs support; these are duplicate fixes for the same code path." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded and conservative overall. The only accepted duplicate pair is well supported by the packet: both PRs modify `src/transformers/loss/loss_utils.py`, target issue 43240, and implement the same `fixed_cross_entropy` kwargs expansion. The other soft-edge pairs are rejected in a cautious, file- and change-based way without overclaiming a shared bug." + }, + "evaluator_used": true, + "retried": false, + "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": "cc55c2bca5fcae75723397ff187f7da679e99a1eb725387e982aee32259b7fcb", + "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": "These are related VibeVoice PRs, but they are not duplicates: PR 40546 is the base VibeVoice implementation, while PR 42781 adds a separate realtime variant with acoustic-tokenizer support and additional generation/feature-extraction work.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 40546 is the better canonical PR because it is the broader, foundational VibeVoice implementation and already covers the core model integration, docs, and Auto-class wiring.", + "best_issue_reason": null, + "best_pr_reason": "PR 40546 is the best representative because it establishes the main VibeVoice codepath; PR 42781 is a later draft focused on realtime-specific extensions rather than the same concrete change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "They overlap on VibeVoice plumbing, but 42781 is a distinct realtime follow-up with extra acoustic-tokenizer and generation changes. It is not the same underlying fix/change and would not be safely 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": 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": "da5861b5337a1a214beebad06de6747884ec825e5e5e829d4975ef0e399c0a51", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4675, + "estimated_input_tokens": 1169, + "estimated_eval_tokens": 2594 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two pull requests are unrelated feature additions: one adds covariate-based forecasting to TimesFM, the other adds DINOv3 image classification support. They do not appear to fix the same code-path problem or be mergeable into one PR.", + "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:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different models, different subsystems, and different goals: TimesFM covariate forecasting vs. DINOv3 image classification. No shared underlying bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "6629ed434b5547738597a9e45f3ae5335e5c39796b1266be3a0ae38a01bff6f7", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10408, + "estimated_input_tokens": 2602, + "estimated_eval_tokens": 5460 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10408, + "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 independent model-addition PRs (DeepSeek OCR, Isaac, and Kimi2-6). They share only the standard Transformers model-onboarding scaffolding, not the same underlying change.", + "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": "Different model integrations: DeepSeek OCR vs Isaac. Shared files are generic registry/docs/template updates typical of new model PRs, not the same code-path fix or mergeable duplicate change." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "Different model integrations: Isaac vs Kimi2-6. The overlap is broad onboarding machinery and docs, but the concrete model code, configs, and tests are for unrelated models, so these are not duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "b9ddd2a1be960ca3b58f11ff5f827b7d5486751a47a888b378a036a7264e13a4", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related only at the level of GitHub Actions maintenance, but they target different action families and different changes: PR 42975 upgrades core GitHub actions for Node 24 compatibility, while PR 42976 updates docker/build-push-action to v6 across Docker image workflows. They should not be merged as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 42975 is the clearer canonical representative: it has the more specific stated goal (Node 24 compatibility) and a broader, more coherent set of workflow action upgrades.", + "best_issue_reason": null, + "best_pr_reason": "PR 42975 is the best overall representative of the cluster because its scope is more self-contained and explicitly framed, whereas PR 42976 is a separate Docker action version bump.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Different underlying changes: one upgrades core GitHub Actions for Node 24 compatibility, the other bumps docker/build-push-action to v6 in Docker image workflows. Same general maintenance area, but not the same fix/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "443824e7b861755b7db7690480fdf0858e946a73fac77320496b56b73d50fbbf", + "component_budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 9, + "serialized_chars": 23541, + "estimated_input_tokens": 5886, + "estimated_eval_tokens": 12028 + }, + "budget": { + "node_count": 7, + "item_count": 7, + "soft_pair_count": 9, + "serialized_chars": 23541, + "estimated_input_tokens": 5886, + "estimated_eval_tokens": 12028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a loose topical grouping of unrelated PRs: several independent model-addition PRs (ViT NEPA, Molmo, Molmo2, Lfm2 Audio, CTSM, PP-OCRv6 support) plus one distributed pipeline-parallel infrastructure PR. The apparent overlap is mostly boilerplate docs/auto-registry changes, not duplicate code changes.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR 43451 is the strongest representative item here: it is a large, highly reviewed model-addition PR with substantial implementation/test coverage. It is not a duplicate of the others, but it is the most developed single PR in the cluster.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model families and code paths: ViT NEPA vs Molmo. Shared docs/auto-registration edits are standard boilerplate for model adds, not the same change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "ViT NEPA and Molmo2 are separate model additions with different implementations, processors, and tests; no shared bug or identical feature change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "ViT NEPA vs CTSM are unrelated model additions. Similar registry/docs edits do not indicate the same underlying change." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are related by name but are distinct models with different files and functionality; this is not one fix/feature split across two PRs." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM are different model stacks and different changes entirely." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM are separate model additions with no shared concrete code-path problem." + }, + { + "left": "pull_request:43973", + "right": "pull_request:45838", + "accept": false, + "reason": "Lfm2 Audio adds an audio model, while PP-OCRv6 adds text-recognition models. Different modalities, code paths, and goals." + }, + { + "left": "pull_request:44601", + "right": "pull_request:45838", + "accept": false, + "reason": "Pipeline-parallel loading support is core distributed infrastructure; PP-OCRv6 is a model-family addition. They are not the same bug or change." + }, + { + "left": "pull_request:45490", + "right": "pull_request:45838", + "accept": false, + "reason": "CTSM and PP-OCRv6 are different model-family additions with distinct implementations and tests, not a duplicate pair." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "98ae86de9abe81f55837c73daa31d2407544121d849be89f7be1c0b8f1f86cdd", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7093, + "estimated_input_tokens": 1774, + "estimated_eval_tokens": 3804 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 7093, + "estimated_input_tokens": 1774, + "estimated_eval_tokens": 3804 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These three PRs are not duplicates. 43291 is a Whisper tokenizer/test fix, 43665 is mostly test-only noise, and 44815 is a substantive FP8/model-loading dequant change. The only apparent similarity is broad \u201cfix\u201d/test activity, which is not enough to merge them.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44815 is the clearest representative of the cluster because it contains the only substantial code-path fix; the other two PRs are unrelated test/Whisper changes.", + "best_issue_reason": null, + "best_pr_reason": "PR 44815 is the strongest standalone PR: it fixes core model-loading / fine-grained FP8 behavior and includes targeted related changes, unlike the other two which are unrelated or test-only.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Different areas and different kinds of changes: Whisper tokenizer logic vs CLIP/ViT test edits. No shared bug or mergeable combined fix." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper timestamp decoding/tokenization is unrelated to FP8 dequant/model loading. Same generic \u201cfix\u201d label only." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "43665 is test-only noise for CLIP/ViT, while 44815 fixes FP8 dequant/model loading. Not the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "38c12097a507280dfb1541a78d110ddad94ad595ef053af4c08c70103f26b0c1", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4563, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4563, + "estimated_input_tokens": 1141, + "estimated_eval_tokens": 2538 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two open PRs are not duplicates: they target different changes, and the shared Vit test filename looks incidental. PR 43488 is a small setup/version/test-bot check, while PR 43532 changes CI workflows and testing utilities. Do not merge them together.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "If one PR must be singled out as the cluster representative, 43532 is broader and has more context, but this is not a true duplicate cluster.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "Different underlying changes: 43488 is about setup.py/dependency/version and a Vit test tweak, while 43532 modifies CI workflows, testing utilities, and related test files. Shared test file alone is not enough to treat them as the same 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": 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": "625cd7798fc97138e880b6a7f06ec9f37d583258da591fcc9c2610aeec11d751", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6082, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate pull requests adding different segmentation pipelines: one for Promptable Concept Segmentation and one for Promptable Visual Segmentation. They share some documentation and registration scaffolding, but the concrete pipeline behavior, model coverage, and test targets differ, so they should not be treated as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Different underlying changes: one introduces Promptable Concept Segmentation, the other Promptable Visual Segmentation. Shared docs/auto-registration files are incidental and do not indicate the same code-path fix or a PR that could be merged as one." + } + ] + }, + "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": "60e265cd84b773917a5e3e13cb74aa8d8f067bfbf56046e82da82442e4ba2ef5", + "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 in the same trainer file, but they fix different trainer extension points: one changes gradient-accumulation loss scaling, the other extracts signature-column filtering for overrideability. They are related in subsystem, not duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR in this cluster: the PRs target different code paths and different customization hooks, so they should not be collapsed into one change.", + "best_issue_reason": null, + "best_pr_reason": "PR #43651 is the more representative concrete bug-fix style change in this cluster because it directly alters training-step loss normalization behavior; however, it still is not a duplicate of #43663.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different underlying problems: #43651 adds control over gradient-accumulation loss scaling, while #43663 refactors signature-column inspection for custom trainers/PEFT. Same file and similar theme, but not the same 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": 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": "1b3dc0ba6da7dd1bbec9b2b52eaa64f0839ce5ca8b062e4c7194cbbbf6b9f2d4", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5991, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 5991, + "estimated_input_tokens": 1498, + "estimated_eval_tokens": 3252 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs cluster around a CLI `serve.py` annotation/runtime NameError issue involving `TypeAdapter` and related typing cleanup. The strongest duplicate pair is the two explicit `TypeAdapter`-focused fixes; the more generic annotation cleanup looks related but not clearly the same concrete change.", + "confidence": 0.67, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43842 is the most precise and narrowly scoped fix for the reported `TypeAdapter` NameError when `pydantic` is absent, with a minimal change set in the affected code path.", + "best_issue_reason": null, + "best_pr_reason": "PR #43842 is the best representative because it directly targets the specific runtime failure, uses the smallest focused patch, and is least likely to include unrelated typing cleanup.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Both touch `src/transformers/cli/serve.py` and relate to annotation-name errors, but #43656 is broader future-annotations/type-hint cleanup while #43842 specifically fixes `TypeAdapter` availability; not clearly the same concrete bug fix." + }, + { + "left": "pull_request:43836", + "right": "pull_request:43842", + "accept": true, + "reason": "Both are clearly aimed at the same `TypeAdapter` NameError in `serve.py`, using stringized annotations/TYPE_CHECKING to avoid runtime evaluation; these look like overlapping fixes to the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The clustering is grounded in the packet. The distinction between 43656 and 43842 is conservative, and the 43836/43842 pairing is reasonably supported by the shared `serve.py` TypeAdapter fix and shared issue target. The summary does not overclaim beyond the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "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:44161", + "pull_request:44017|pull_request:44722", + "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:44073|pull_request:44161", + "pull_request:44059|pull_request:44068", + "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": "c7f05932f627e69748760f023cc4a63ce77ec28a38aee79e101cf9363479dc37", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84891, + "estimated_input_tokens": 21223, + "estimated_eval_tokens": 42702 + }, + "budget": { + "node_count": 27, + "item_count": 27, + "soft_pair_count": 72, + "serialized_chars": 75088, + "estimated_input_tokens": 18772, + "estimated_eval_tokens": 37800 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a batch of model-specific refactors for standardized output tracing around issue #43979. The only clear duplicate-like overlaps are within the GPT-Neo pair and the GPT-J/CodeGen sequence; the rest are separate model refactors that share a theme but not the same concrete change.", + "confidence": 0.82, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44722 is the best canonical representative for the concrete GPT-J/CodeGen output-tracing refactor: it has the clearest implementation scope, explicit issue linkage, and subsumes the closely related GPT-J-only variants.", + "best_issue_reason": null, + "best_pr_reason": "PR 44722 is the strongest single PR in this cluster because it most clearly captures the underlying GPT-J/CodeGen output-tracing change and is the most complete version of that specific refactor.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families and code paths; same issue target is not enough to make these duplicates." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "CVT/FNet output tracing vs DeBERTa output tracing are separate model refactors." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Both touch the GPT-J output-tracing refactor on the same code path; 44066 just includes an added CodeGen companion change." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "Different vision model refactors (ResNet/RT-DETR ResNet vs EfficientNet), not the same concrete change." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet output tracing and GPT-J/CodeGen output tracing are unrelated concrete changes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen refactor vs EfficientNet refactor; only the shared theme is standardized tracing." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa are unrelated model implementations with different refactor targets." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen output tracing and MGP-STR output tracing are different code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen refactor vs MPT refactor; same theme, different model behavior." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "These are both GPT-Neo output-tracing refactors of the same model code path and look like the same underlying change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen and GPT-Neo are separate model refactors, so this is only thematic similarity." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet output tracing and GPT-J/CodeGen output tracing are unrelated changes." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are different model-specific refactors." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet and GPT-J are unrelated code paths despite the shared tracing theme." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are different model implementations with distinct output handling." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet vs MPT is not the same underlying bug/change." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo output tracing and Flaubert output tracing are separate model changes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are distinct model refactors, not one duplicate." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are different concrete output-tracing changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are separate models and separate changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo refactor vs MGP-STR refactor; same pattern, different code." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model refactors." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert output-handling changes are unrelated concrete changes." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT2 and MGP-STR output tracing are separate code paths." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are unrelated model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT2 refactor and GPT-J/CodeGen refactor are not the same bug/change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo vs MPT; only the broad tracing theme is shared." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are different model-specific output-tracing refactors." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer and GPT-J/CodeGen are unrelated concrete changes." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are separate models with different implementation details." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44076", + "accept": false, + "reason": "MPT and ImageGPT are different model refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44073", + "accept": false, + "reason": "RegNet and VisualBert are unrelated model-specific changes." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44071", + "accept": false, + "reason": "SegFormer vs MPT is only a shared architecture pattern, not the same change." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44101", + "accept": false, + "reason": "VisualBert and Flaubert are different models and different output tracing updates." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT2 and MPT changes are unrelated concrete refactors." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "MPNet and LongT5 output tracing are separate code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44018", + "accept": false, + "reason": "RegNet and GPT-Neo are unrelated implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44056", + "accept": false, + "reason": "SegFormer and MPNet are different model refactors." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBert and MPT are different concrete changes." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44161", + "accept": false, + "reason": "VisualBert and LongT5 are separate model-specific output-tracing refactors." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44056", + "accept": false, + "reason": "RegNet and MPNet are unrelated code paths." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 are not the same underlying bug/change." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44161", + "accept": false, + "reason": "TextNet and LongT5 are different model refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT2 are separate model-specific changes." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44071", + "accept": false, + "reason": "MPNet and MPT are different implementations and not duplicates." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44018", + "accept": false, + "reason": "SqueezeBert and GPT-Neo are unrelated concrete changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and SegFormer are different model refactors." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBert and DeBERTa are unrelated output-tracing changes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTa and TextNet are separate model-specific changes." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBert and VisualBert are different code paths." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are separate model refactors." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and VITS are different concrete output-handling changes." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44068", + "accept": false, + "reason": "MPNet and GPT-Neo are unrelated model refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different concrete changes." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: most PRs are distinct model-specific output-tracing refactors, and the GPT-Neo and GPT-J/CodeGen overlaps are the only clear mergeable duplicates. The chosen canonical PR rationale is also supported by the diffs and issue linkage. Soft-edge verdicts are mostly conservative and do not overstate duplication." + }, + "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": "0869a19b82fbb11817c2090a714046ed5347e7b46aae7422b0a476f28958921a", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84891, + "estimated_input_tokens": 21223, + "estimated_eval_tokens": 42702 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 6, + "serialized_chars": 17124, + "estimated_input_tokens": 4281, + "estimated_eval_tokens": 8818 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These PRs are all instances of the same broad refactor pattern\u2014migrating individual model forwards to standardized output-capture utilities\u2014but they target different model implementations and code paths. As a duplicate cluster, they should stay separate; none of the soft edges are true duplicates under a strict PR triage standard.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44129 is the strongest cluster representative because it has the clearest problem statement, an explicit issue link, and a substantial output-tracing refactor in a model file consistent with the rest of the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR #44129 is the best single PR to represent the cluster since it most explicitly frames the standardized output-capture refactor and is easier to map back to the underlying initiative than the more model-specific variants.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "Same refactor theme, but different models and different code paths. These would not plausibly be merged into one PR without becoming an unrelated multi-model sweep." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Both are output-tracing refactors, but they touch different architectures and files. Similar mechanics are not enough to make them duplicates." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "MPNet and VisualBert are separate model implementations; the shared decorator/capture approach is a framework pattern, not the same underlying bug or change." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "Distinct model-specific refactors with no shared concrete defect or identical code path." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are unrelated model files. This is the same migration style, not the same change." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "These target different model internals and do not fix the same concrete issue; they only share the standardized output-capture motif." + } + ] + }, + "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": "55567ac97fae8f84953354214a32da6800b280f4fe44ad1b21474baebfb34a32", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40172, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 40, + "serialized_chars": 40172, + "estimated_input_tokens": 10043, + "estimated_eval_tokens": 20342 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs share a common refactor theme, but they are separate model-specific changes rather than duplicates of one underlying bug or patch. I would not merge any soft edge pairs.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44003 is the most representative cluster anchor: it is a broader output-tracing refactor, spans two closely related model files, and introduces shared tracing helpers rather than a one-off local tweak.", + "best_issue_reason": null, + "best_pr_reason": "PR 44003 is the best overall representative because it is the broadest and most infrastructure-oriented of the set, making it a better canonical example than the single-model refactors.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families and files; same refactor theme only, not the same concrete bug/change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different models/code paths; only the output-tracing refactor pattern overlaps." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Unrelated model implementation; no shared underlying fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Different architectures and code paths; not one mergeable fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families; same tracing refactor motif but no duplicate bug." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "Separate models with different forward paths; not the same change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different subsystems and model code paths; only a generic refactor similarity." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families and filenames; not the same concrete fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Encoder-decoder wrapper vs MobileNet model; no shared underlying bug." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "Different model code paths; same refactor pattern only." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Different models and output paths; not a duplicate change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different architectures and files; no same concrete bug/change." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "Distinct model families; too broad a similarity to merge." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model implementations and code paths; only same refactor style." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "Separate model code paths; not the same underlying issue." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Mamba/FalconMamba vs UnivNet; unrelated concrete changes." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "Different models and outputs; no shared bug to coalesce." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and forward implementations; not duplicates." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Distinct code paths in different architectures; same meta-theme only." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families and files; not the same fix." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Mamba/RWKV vs CodeGen are unrelated code paths; do not merge as duplicates." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model subsystems; only the tracing refactor label matches." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "Separate model implementations; not one concrete bug." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model families and forward logic; not duplicates." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44004", + "accept": false, + "reason": "Vision/text dual encoder vs CodeGen are unrelated code paths." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Different model families; similar refactor pattern only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "DepthAnything vs SpeechEncoderDecoder are unrelated concrete changes." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Different models and output structures; not the same fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures and files; no shared bug." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "DepthAnything vs DPR are distinct changes in different code paths." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Different encoder-decoder wrappers vs RWKV; not one underlying issue." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Different model wrappers and files; same tracing pattern only." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different models and concrete forward paths; not duplicates." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "Vision/audio encoder-decoder vs DPR are unrelated code changes." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "Different models and implementations; no shared bug/change." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Speech encoder-decoder and RWKV are distinct code paths." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "Different models and output mechanisms; not mergeable as duplicates." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint vs RWKV are unrelated changes despite similar refactor wording." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model families and code paths; not the same fix." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are different implementations; only the output-tracing refactor theme overlaps." + } + ] + }, + "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": "f924fe59d79e0dce6c7f557453056dd024ef64ccba3a6dc6bde8c7912eef62d9", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10913, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10913, + "estimated_input_tokens": 2729, + "estimated_eval_tokens": 5714 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate model-addition PRs (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They share some Transformers registry/docs plumbing, but they do not appear to fix the same underlying bug or implement the same concrete change, so none should be deduplicated together.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45075 is the best cluster representative: it is a substantial, non-draft model integration with conversion code, modeling, processing, tests, and docs. The other PRs are distinct model additions and not duplicates.", + "best_issue_reason": null, + "best_pr_reason": "PR #45075 is the strongest standalone PR in this set because it looks the most complete and self-contained, with explicit conversion and broad implementation coverage.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "PenguinVL and Voxtral TTS are different models with different codepaths; overlap is limited to shared Transformers registry/docs files, which is not enough to treat them as the same change." + }, + { + "left": "pull_request:44662", + "right": "pull_request:45075", + "accept": false, + "reason": "PenguinVL and DeepSeek-OCR-2 are unrelated model implementations. Shared auto-registration and docs files are expected for separate additions, not evidence of duplication." + }, + { + "left": "pull_request:45075", + "right": "pull_request:45401", + "accept": false, + "reason": "DeepSeek-OCR-2 and Voxtral TTS target different architectures and functionality. They only share common model-registration plumbing, so they should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "e30cbd0e07823ba3b76575d557f9fee68a9e9a003e77cfafdae5078174e677cf", + "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": "The two PRs are thematically related around fusion, but they target different layers: one introduces a general module-fusion API, while the other wires up n-to-1 kernel fusion through KernelConfig and model-loading integration. They are not duplicates of the same concrete change.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45363 is the more concrete end-user feature: it connects KernelConfig to model loading and kernel-backed fusion across multiple integration points. It is the better representative for the cluster\u2019s practical fusion functionality.", + "best_issue_reason": null, + "best_pr_reason": "PR #45363 is the strongest overall representative because it has the clearest feature surface, explicit integration changes, and a user-visible workflow. PR #44979 is more of a foundational API piece than the full feature.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Related area, but not the same underlying change. #44979 adds a standalone module fusion API in src/transformers/module_fusion.py; #45363 implements n-to-1 kernel fusion via KernelConfig and loader/integration plumbing. Different abstractions and code paths, so they should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "4e3706aa6b9f0a8b7550d2da59a2e3e3f9abe18701f80fce9bbc8879c79d92e9", + "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 fix the same AttributeError in `_patch_mistral_regex` by switching from `tokenizer.backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; 45317 is the more complete version because it includes tests and clearer framing.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "45317 is the better canonical PR because it targets the same bug as 45086 but adds test coverage and a clearer description of the fix.", + "best_issue_reason": null, + "best_pr_reason": "45317 is the best representative PR for this cluster since it fully addresses the bug and includes validation, whereas 45086 is a smaller patch of the same change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying bug and same code path in `tokenization_utils_tokenizers.py`; the later PR is essentially a superset with added tests, so they could plausibly be merged into one fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "7f8ced45f61375ec7d764c00d673fdfe23dc51164b968d0bde7e31ce10247fc6", + "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": "Related NemotronH work, but not duplicates: PR #45115 is a conversion/refactor for legacy NemotronH weight mapping and GraniteMoeHybrid-compatible naming, while PR #45569 is a broader feature split that adds NemotronH Dense/Sparse variants, auto-mapping updates, docs, and tests.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45569 is the broader umbrella change and the better representative of the NemotronH family update; it includes the new Dense/Sparse model structure and the surrounding integration work.", + "best_issue_reason": null, + "best_pr_reason": "PR #45569 is the most complete and user-visible change in the cluster, so it is the best single PR to anchor on if one must be chosen.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "They touch some of the same NemotronH files, but the underlying changes are different: #45115 is a conversion/mapping refactor for inheritance from GraniteMoeHybrid, while #45569 introduces Dense/Sparse model variants plus docs/tests. They would not plausibly merge 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": 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": "ec39b6b427c5404d80288dd59f606316fdede2caff9acbc8baac59d335402ae4", + "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": "Two related PRs in the heterogeneous-model workstream, but they are not the same change: #45333 adds per-layer heterogeneous config support, while #45332 builds on that with broader modeling/cache changes for heterogeneous models. They should stay separate.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45332 is the broader umbrella change: it adds heterogeneous model support across cache/modeling code and includes the downstream consumers of the new per-layer config abstractions.", + "best_issue_reason": null, + "best_pr_reason": "PR #45332 is the best representative of the cluster because it spans the full heterogeneous-model feature set, not just the config plumbing.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Related work, but not the same concrete change. #45333 is config plumbing only; #45332 extends heterogeneous support into modeling/cache behavior and is broader than a mergeable duplicate." + } + ] + }, + "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": "ef2d5f67b781a3ee50ac25c5911b63a0b1e43e5b3ae727afe5ed9244cde250ec", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5514, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs are typing-related, but they address different scopes: #45415 is a broad repository-wide type-checking/configuration pass, while #45425 adds specific typing protocols and helpers for model/PEFT-related code paths. They are related, but not the same change.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45425 is the more concrete and representative typing change: it adds specific protocol/type definitions used by model-related code, rather than a broad type-checking sweep.", + "best_issue_reason": null, + "best_pr_reason": "PR #45425 is the best single representative artifact because it has a focused implementation goal and clear code-path impact, whereas #45415 is a wider type-checking cleanup across many files.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Do not treat as duplicates: #45415 is a broad type-checking pass/config change across many modules, while #45425 introduces specific typing helpers for model/PEFT integration code. Related theme, different underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "70f362b2fd0f75bfa7df5896e8e4a2b1e4ba3dc7143168794f4318e6bf2763c9", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 6, + "serialized_chars": 14704, + "estimated_input_tokens": 3676, + "estimated_eval_tokens": 7608 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 6, + "serialized_chars": 14704, + "estimated_input_tokens": 3676, + "estimated_eval_tokens": 7608 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are mostly separate model-specific fixes or test expectation updates. They share a broad theme of parallel/XPU-related work, but not the same underlying bug/change, so none of the soft edges should be merged.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45817 is the strongest representative: it contains a concrete code fix for Gemma4 model-parallel behavior, with an actual implementation change rather than only test-data updates.", + "best_issue_reason": null, + "best_pr_reason": "PR 45817 is the best single PR to represent the cluster because it is a substantive bug fix in model-parallel logic and is more central than the small test-only PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45487", + "right": "pull_request:45843", + "accept": false, + "reason": "Both are model-parallel fixes, but for different models and different failure modes: AltCLIP/ChineseCLIP device placement vs Deimv2 _no_split_modules propagation. Not the same underlying bug." + }, + { + "left": "pull_request:45487", + "right": "pull_request:45926", + "accept": false, + "reason": "AltCLIP/ChineseCLIP code fix versus HyperCLOVAX XPU expectation updates. Different models and one is test-only, so they should not be merged." + }, + { + "left": "pull_request:45817", + "right": "pull_request:45843", + "accept": false, + "reason": "Gemma4 and Deimv2 have distinct model-parallel code paths and fixes. Same broad area, but not the same concrete bug." + }, + { + "left": "pull_request:45858", + "right": "pull_request:45890", + "accept": false, + "reason": "Both only add XPU expectations, but they target different model test suites and different outputs. No shared underlying bug/change." + }, + { + "left": "pull_request:45858", + "right": "pull_request:45926", + "accept": false, + "reason": "Both are test expectation updates, but for different models and test cases. Too broad to treat as duplicates." + }, + { + "left": "pull_request:45890", + "right": "pull_request:45926", + "accept": false, + "reason": "These are separate XPU expectation updates for different model suites, not the same concrete fix 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": "d2a708fe0009cc9081a1359d6893ba957a0123cdcc1ce169989d2d9b2f4304fa", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two PRs both touch expert parallelism, but they address different failures: one is about preserving/loading EP-related config and enabling EP detection during DeepSpeed ZeRO-3 loading, while the other fixes FSDP2/rank-0 broadcast overwriting expert parameters during distributed loading. They are related subsystem-wise but not the same underlying bug or change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR 45662 is the more concrete and self-contained bug fix: it targets a specific expert-parameter overwrite path during distributed loading and includes code changes directly addressing that failure mode.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Different concrete bugs and code paths: 45548 focuses on config serialization/EP detection for DeepSpeed ZeRO-3 loading, while 45662 fixes expert tensors being overwritten during FSDP2/loading broadcast. They should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "ad207c371f962bbdddaf4437f9f0e89b21e205efadafb7dbf26f966a21593228", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4590, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4590, + "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 address the same xdist-related captured_info/debug-log collision problem in testing_utils; 45645 is the more complete variant because it also updates CI log collection to handle per-worker files.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45645 is the best canonical representative because it includes the core xdist-safe testing_utils fix plus the downstream CI workflow adjustment needed to consume multiple captured_info files.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the strongest standalone candidate: it covers the same underlying bug as 45639 and adds the supporting CI handling, making it the more complete fix.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Accept: both PRs fix the same captured_info/xdist collision in the test debugging path, and 45645 looks like a broader follow-up of the same concrete change rather than a different bug." + } + ] + }, + "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-45851-2", + "nodes": [ + "pull_request:45851", + "pull_request:45947" + ], + "soft_pairs": [ + "pull_request:45851|pull_request:45947" + ], + "prepared_review_unit_hash": "2133b4f90d7396d57ff929fb3a603b0289caf70ccdb2be432cae7e415eb494d1", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4105, + "estimated_input_tokens": 1027, + "estimated_eval_tokens": 2310 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4105, + "estimated_input_tokens": 1027, + "estimated_eval_tokens": 2310 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open docs PRs in the chat templating area, but they address different topics: template file storage/loading vs. reasoning-field prefill behavior. They should not be deduplicated.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45851 is the better anchor because it is explicitly linked to issue #45205 and has a focused, self-contained docs change about chat template storage/loading.", + "best_issue_reason": null, + "best_pr_reason": "PR #45851 is the strongest representative: it has an explicit issue target, a clearer scoped documentation fix, and is more substantively connected to a concrete user-facing gap than #45947.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45851", + "right": "pull_request:45947", + "accept": false, + "reason": "They are both chat-templating docs, but they describe different changes and different behaviors. #45851 explains on-disk chat template shapes; #45947 documents prefilling a reasoning field instead of content. 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 + } + ] +} diff --git a/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/manifest.json b/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..7b180c36859e72fa43f365651f6e54c5a9e87b5f --- /dev/null +++ b/snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260514t000044z", + "artifacts": { + "hybrid": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-05-14T00:09:20Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260514T000044Z", + "variant": "hybrid" +} diff --git a/snapshots/20260514T000044Z/manifest.json b/snapshots/20260514T000044Z/manifest.json index 0f9e6b2ffecdce1628e1a719f5721dbc8e0448d7..d79264c6ecd3c02ffc1e82581b499e5bbe9c4d62 100644 --- a/snapshots/20260514T000044Z/manifest.json +++ b/snapshots/20260514T000044Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 656 }, "extracted_at": "2026-05-14T00:00:44Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260514t000044z", + "runs": { + "hybrid-model-20260514t000044z": { + "analysis_id": "hybrid-model-20260514t000044z", + "artifacts": { + "hybrid": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260514T000044Z/analysis-runs/hybrid-model-20260514t000044z/manifest.json", + "model": null, + "published_at": "2026-05-14T00:09:20Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260514T000044Z", "watermark": {