diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index d50f7fda1256d5e8ddd32d0bd57f09aee5c319ad..f5eda7d1a044a858465787cb3327a2a70f7372db 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": "20260526T060025Z", - "generated_at": "2026-05-26T06:16:50Z", + "snapshot_id": "20260526T120027Z", + "generated_at": "2026-05-26T12:14:03Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ @@ -1755,64 +1755,6 @@ } ] }, - { - "cluster_id": "cluster-44001-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 44004, - "issue_numbers": [], - "pr_numbers": [ - 44001, - 44004 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 44001, - "right_pr_number": 44004, - "code_similarity": 0.087, - "size_similarity": 0.237, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.262, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, - { - "cluster_id": "cluster-44002-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 44002, - "issue_numbers": [], - "pr_numbers": [ - 44002, - 44029 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 44002, - "right_pr_number": 44029, - "code_similarity": 0.135, - "size_similarity": 0.43, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.329, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, { "cluster_id": "cluster-44018-2", "summary": "Cluster of 2 related pull requests linked by soft_similarity.", @@ -2144,24 +2086,6 @@ "target_issue_number": 43979, "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." }, - { - "cluster_id": "cluster-44001-2", - "canonical_pr_number": 44004, - "duplicate_pr_numbers": [ - 44001 - ], - "target_issue_number": null, - "reason": "PRs in cluster-44001-2 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-44002-2", - "canonical_pr_number": 44002, - "duplicate_pr_numbers": [ - 44029 - ], - "target_issue_number": null, - "reason": "PRs in cluster-44002-2 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-44018-2", "canonical_pr_number": 44068, @@ -2220,8 +2144,8 @@ "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 + "reason": "Issue #45647 is the strongest global issue candidate because it is open, belongs to a cluster with 3 artifacts, and carries 13 discussion comments plus 8 inbound references.", + "score": 164.0 }, "best_pr": { "cluster_id": "cluster-43979-11", diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index e3812fc726a0e890ae98b10ff1ae4567f137a964..bcea7d1543c411f4fd2252321cbe8d13fd1a0679 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": "20260526T060025Z", - "generated_at": "2026-05-26T06:16:50Z", + "snapshot_id": "20260526T120027Z", + "generated_at": "2026-05-26T12:14:03Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -110,8 +110,8 @@ "issue:43828|issue:45161", "issue:44918|issue:45161", "issue:43828|issue:44292", - "issue:43541|issue:43716", "issue:43716|issue:45685", + "issue:43541|issue:43716", "issue:44292|issue:44918", "issue:43716|issue:44292", "issue:45507|issue:45685", @@ -123,8 +123,8 @@ "issue:44918|issue:45685", "issue:43039|issue:43828", "issue:30277|issue:36331", - "issue:45237|issue:46015", "issue:34928|issue:36010", + "issue:45237|issue:46015", "issue:43039|issue:44805", "issue:41762|issue:43596", "issue:39290|issue:41720", @@ -138,412 +138,364 @@ "issue:43866|issue:44863", "issue:44292|issue:46015" ], - "prepared_review_unit_hash": "bc6216ccb136b933f6d89ef1e2f787562616bd8499112b53613f6d4a0d74143e", + "prepared_review_unit_hash": "104455d95fac3691ee93543a2b57e7f7838dca6572c1fe44d13cdd596cb0f895", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 69, - "serialized_chars": 48396, - "estimated_input_tokens": 12099, - "estimated_eval_tokens": 24454 + "serialized_chars": 48407, + "estimated_input_tokens": 12102, + "estimated_eval_tokens": 24460 }, "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 that only share broad subsystem themes (DeepSpeed, tokenization, device mapping, docs). I only see two true duplicate-like pairs: the BERT/DeepSpeed ZeRO-3 index-out-of-bounds reports and the Granite AutoTokenizer token-ID regression reports. The rest should stay separate.", - "confidence": 0.84, - "canonical_issue_reason": "Issue 45920 is the broadest and most current report in the tokenization-regression subcluster; it generalizes the Granite-specific 45812 bug to multiple model families, making it the best representative issue here.", + "summary": "This cluster is mostly noisy lexical similarity across unrelated issues. The only clear duplicate pockets are the DeepSpeed ZeRO-3 zero-sized-tensor index error reports and the AutoTokenizer wrong-token-ID regression; the rest should stay separate.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43596 is the best representative of the genuine duplicate pocket: it states the generic DeepSpeed ZeRO-3 zero-sized-tensor IndexError clearly, and the near-identical 43638 plus related 41762 reports orbit the same failure mode.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45920 is the strongest global issue candidate: open, broad in scope, and it subsumes a narrower duplicate report while describing a concrete user-facing regression.", + "best_issue_reason": "issue:43596 is the strongest overall anchor for triage because it captures the core ZeRO-3 loading bug in the most general form, with the closest true duplicate nearby; most other items in the cluster are only superficially similar by error wording.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different failures in different paths: config/attribute lookup vs streaming chat API behavior." + "reason": "Different model/config attribute error vs streaming API failure; only broad Qwen similarity." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Both involve backend/device limitations, but the concrete bugs are unrelated (AMD GPU inference vs MPS histogram op on Int)." + "reason": "Different hardware backends and failures (AMD GPU vs MPS histogram op); not the same bug." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Different models and symptoms; dtype mismatch in Mistral-3 is not the same bug as GPT-OSS on AMD GPUs." + "reason": "Different models and symptoms: dtype mismatch in preprocessing vs AMD GPU execution failure." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Both are device/dtype issues, but they affect different models and code paths." + "reason": "Both are runtime failures, but the model, backend, and error class differ." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different model families and failures: AMD GPU runtime issue vs GraniteMoEHybrid invalid method call." + "reason": "Different MoE/runtime problems; no shared concrete code path." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error is a different concrete bug from GPT-OSS AMD GPU failure." + "reason": "Both involve Qwen/GPT-OSS style models, but the reported failures are unrelated." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "DeepSpeed collective mismatch and ONNX export failure are unrelated code paths." + "reason": "DeepSpeed collective mismatch is unrelated to ModernBERT ONNX export." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "ONNX export error vs FineGrainedFP8Config failure; same broad area, not the same bug." + "reason": "ONNX export error and FineGrainedFP8 failure are separate model/tooling issues." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "Different problems: ModernBERT ONNX export vs Qwen3 auto device mapping CUDA assert." + "reason": "ModernBERT export and Qwen3 auto device mapping are unrelated." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Table QA runtime crash and outdated RAG docs are unrelated." + "reason": "Table QA crash and RAG docs cleanup are not the same underlying issue." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are v5 docs cleanup issues, but they refer to different removed docs/features and not one underlying bug." + "reason": "Both are v5 docs problems, but they cover different outdated examples and should be tracked separately." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ONNX export failure and GenerationMixin import error are distinct regressions." + "reason": "ONNX export error vs import error for GenerationMixin; different code paths." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Generic mask shape mismatch and Gemma4 device-mapping mismatch are different bugs." + "reason": "Mask/index shape mismatch is unrelated to Gemma4 device_map forward mismatch." }, { "left": "issue:44918", "right": "issue:46165", "accept": false, - "reason": "Qwen3.5 embedding/trainer failure is unrelated to resume-training with trust_remote_code." + "reason": "TRl SFT embedding unpacking and resume-training/trust-remote-code are different training workflows." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both mention index/shape problems, but the model/setup and failure mode differ." + "reason": "Similar IndexError wording, but one is ZeRO-3/Bert loading and the other is a mask-shape mismatch." }, { "left": "issue:43596", - "right": "issue:44805", - "accept": false, - "reason": "DeepSpeed ZeRO-3 Bert index error is not the same as the generic mask shape mismatch." - }, - { - "left": "issue:44805", - "right": "issue:45161", + "right": "issue:46165", "accept": false, - "reason": "Different model/runtime contexts; generic tensor shape error vs GPT-OSS tensor-parallel issue." + "reason": "Both mention training/loading, but the actual failures are unrelated." }, { "left": "issue:43572", "right": "issue:46165", "accept": false, - "reason": "Pad token config regression and training-resume/trust_remote_code issue are unrelated." + "reason": "StableLmConfig pad token regression is unrelated to resume-training behavior." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice_preset bug and ModernBERT export bug have nothing in common beyond being model-related." + "reason": "BarkProcessor voice_preset issue has nothing to do with ONNX export." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "DeepSpeed collective mismatch is a different failure from the Bert/ZeRO-3 index error." + "reason": "Both involve DeepSpeed, but the collective mismatch and ZeRO-3 IndexError are different bugs." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same DeepSpeed ZeRO-3 + BERT index-out-of-bounds bug; the two reports describe the same underlying failure with only wording/model-suffix differences." + "reason": "Same ZeRO-3 IndexError on zero-sized tensors during Bert loading; wording differs, failure mode is the same." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger kernel cross-entropy dispatch and GPT-OSS tensor-parallel failure are unrelated." + "reason": "Liger cross-entropy invocation and GPT-OSS tensor-parallel failure are different issues." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "DeepSeek detokenization regression and LoRA merge/save inconsistency are different bugs." + "reason": "Incorrect detokenization and LoRA save/reload mismatch are separate regressions." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Qwen-3 NVFP4 inference error and LoRA save/reload mismatch are unrelated." + "reason": "NVFP4 runtime error and LoRA round-trip mismatch do not share a concrete bug." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Activation-checkpointing/FSDP recomputation mismatch is not the same as the mask shape error." + "reason": "Both are tensor-shape-related, but they involve different training stacks and error mechanisms." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Both involve DeepSpeed, but the concrete runtime errors and code paths differ." + "reason": "DeepSpeed weight dimension error and collective mismatch are not the same underlying bug." }, { "left": "issue:45859", "right": "issue:45902", "accept": false, - "reason": "Same class and device/materialization area, but one is CPU-offload incompatibility and the other is uninitialized-memory after to_empty(cuda); not clearly the same bug." + "reason": "Same class name, but CPU-offload incompatibility vs uninitialized-memory after meta materialization are distinct failures." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "Bert/ZeRO-3 index error and GPT-OSS TP issue are unrelated." + "reason": "ZeRO-3 Bert loading and GPT-OSS tensor-parallel behavior are unrelated." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "Activation-checkpointing recomputation mismatch and compute_loss signature change are different regressions." + "reason": "Activation-checkpointing tensor-size mismatch and compute_loss signature change are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 Bert IndexError is unrelated to GPT-OSS tensor-parallel failure." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Different models and failures: Mistral-3 dtype mismatch vs GraniteMoEHybrid invalid method." + "reason": "Different model families and different dtype/method errors." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Gemma3 config attribute error and Mistral regex patch crash are unrelated." + "reason": "Gemma3 config attribute missing and Mistral tokenizer regex patch crash are unrelated." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Both are Gemma4-related, but one is device placement and the other is reverse-embedding lookup cost." + "reason": "Device placement mismatch and expensive reverse embedding lookup are different forward-path problems." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Torch version incompatibility and Qwen-3 NVFP4 runtime failure are different issues." + "reason": "Torch version incompatibility and NVFP4 runtime error are not the same bug." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "Different model stacks and symptoms: FSDP checkpointing mismatch vs CUDA assert on Qwen3 auto device mapping." - }, - { - "left": "issue:43828", - "right": "issue:45161", - "accept": false, - "reason": "Autocast dtype mismatch and GPT-OSS tensor-parallel failure are not the same bug." - }, - { - "left": "issue:44918", - "right": "issue:45161", - "accept": false, - "reason": "Qwen3.5/TRL embedding unpacking issue is unrelated to GPT-OSS TP failure." - }, - { - "left": "issue:43828", - "right": "issue:44292", - "accept": false, - "reason": "Both are inference runtime issues, but they affect different models and different failure mechanisms." - }, - { - "left": "issue:43541", - "right": "issue:43716", - "accept": false, - "reason": "Torch dynamo tracing failure in Mixtral is unrelated to Mistral-3 dtype mismatch." - }, - { - "left": "issue:43716", - "right": "issue:45685", - "accept": false, - "reason": "Different backends and bugs: Mistral-3 dtype mismatch vs MPS histogram op not implemented." - }, - { - "left": "issue:44292", - "right": "issue:44918", - "accept": false, - "reason": "Qwen-3 NVFP4 runtime error and Qwen3.5/TRL embedding failure are different issues." - }, - { - "left": "issue:43716", - "right": "issue:44292", - "accept": false, - "reason": "Different models and failure modes; not a duplicate." - }, - { - "left": "issue:45507", - "right": "issue:45685", - "accept": false, - "reason": "GraniteMoEHybrid invalid method and MPS histogram error are unrelated." - }, - { - "left": "issue:43039", - "right": "issue:44918", - "accept": false, - "reason": "Liger kernel cross-entropy dispatch and Qwen3.5 embedding unpacking are different code paths." - }, - { - "left": "issue:41762", - "right": "issue:43638", - "accept": false, - "reason": "Both mention DeepSpeed/zero-3 style failures, but Gemma3 loading and Bert model init are different bugs." + "reason": "FSDP checkpointing mismatch and Qwen3 auto device mapping failure are unrelated." }, { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions from v5 removals, but they concern different removed topics and should stay separate." + "reason": "Both are docs-related, but they cover different removed commands/tasks." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Mistral-3 dtype mismatch and Qwen3-vl StopIteration are unrelated model bugs." + "reason": "Mistral image dtype mismatch and Qwen3-VL video StopIteration are different model-processing failures." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Different models and mechanisms: autocast dtype mismatch vs MPS histogram op limitation." + "reason": "Autocast dtype mismatch and MPS histogram op not implemented are different backend issues." }, { "left": "issue:44918", "right": "issue:45685", "accept": false, - "reason": "Qwen3.5/TRL embedding issue is unrelated to the MPS histogram error." + "reason": "SFT trainer embedding unpacking and MPS MoE histogram failure are unrelated." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Liger kernel cross-entropy and Phi-tiny-MoE autocast mismatch are different problems." + "reason": "Cross-entropy being called via Liger and Phi-tiny-MoE autocast mismatch are different code paths." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed collective mismatch and compute_loss keyword mismatch are unrelated." + "reason": "Collective mismatch in ZeRO-3 and compute_loss signature breakage are unrelated." }, { - "left": "issue:45237", - "right": "issue:46015", + "left": "issue:34928", + "right": "issue:36010", "accept": false, - "reason": "GPT-OSS AMD GPU runtime failure and outdated RAG docs are unrelated." + "reason": "Checkpointing/FSDP tensor mismatch and GenerationMixin import breakage are unrelated." }, { - "left": "issue:34928", - "right": "issue:36010", + "left": "issue:45237", + "right": "issue:46015", "accept": false, - "reason": "FSDP activation-checkpointing mismatch and GenerationMixin import error are different bugs." + "reason": "AMD GPU execution failure and outdated RAG docs are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Liger kernel cross-entropy call and mask shape mismatch are unrelated." + "reason": "Liger cross-entropy call and generic mask-shape IndexError are not the same bug." }, { "left": "issue:41762", "right": "issue:43596", + "accept": true, + "reason": "Same ZeRO-3 zero-sized-tensor IndexError during loading; Gemma3 and Bert are model-specific manifestations of the same failure." + }, + { + "left": "issue:39290", + "right": "issue:41720", "accept": false, - "reason": "Gemma3 ZeRO-3 loading error and Bert ZeRO-3 index error are different model-specific failures." + "reason": "Missing config attribute for vLLM and CUDA assert under auto device mapping are unrelated." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "TextClassificationPipeline docs mismatch and run-command docs removal are separate documentation issues." + "reason": "TextClassificationPipeline docs and removed run-command docs are separate documentation issues." }, { "left": "issue:45237", "right": "issue:45901", "accept": false, - "reason": "GPT-OSS AMD GPU inference failure and table-question-answering crash are unrelated." + "reason": "AMD GPU model failure and table-question-answering crash are unrelated." }, { "left": "issue:44841", "right": "issue:45823", "accept": false, - "reason": "Processor failure for Voxtral and Gemma4 device mismatch are different model/processor bugs." + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are different issues." }, { "left": "issue:43039", "right": "issue:44560", "accept": false, - "reason": "Liger kernel dispatch issue and Qwen3-vl video StopIteration are unrelated." + "reason": "Liger kernel cross-entropy path and Qwen3-VL video StopIteration are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization regression and RAG docs staleness are unrelated." }, { "left": "issue:45812", "right": "issue:45920", "accept": true, - "reason": "Same token-ID regression in AutoTokenizer; 45920 is the broader report that subsumes the Granite-specific 45812 case." + "reason": "Same AutoTokenizer wrong-token-ID regression across different model families; titles point to the same tokenizer bug." }, { "left": "issue:44863", "right": "issue:46015", "accept": false, - "reason": "NemotronH checkpoint-loading bug and outdated RAG docs are unrelated." + "reason": "NemotronH checkpoint loading and RAG docs are unrelated." }, { "left": "issue:43866", "right": "issue:44863", "accept": false, - "reason": "Corrupted checkpoint report and NemotronH loading failure are different issues." + "reason": "Checkpoint corruption and checkpoint loading implementation failure are different problems." }, { "left": "issue:44292", "right": "issue:46015", "accept": false, - "reason": "Qwen-3 NVFP4 inference bug and outdated RAG docs are unrelated." + "reason": "Qwen-3 NVFP4 runtime error and RAG docs cleanup are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The two accepted duplicate-like pairs are well supported by the issue titles and descriptions, and the rejected soft edges mostly distinguish different models, APIs, or failure modes rather than overclaiming duplicates. The representative-issue choice of 45920 over 45812 is also reasonable because it is broader and open." + "feedback": "The summary is well grounded in the packet. The two accepted duplicate pockets are supported by the titles and soft-pair evidence: the DeepSpeed ZeRO-3 zero-sized-tensor IndexError reports (43596/43638/41762) and the AutoTokenizer wrong-token-ID regression (45812/45920). The false soft-edge rejections are conservative and the reasons mostly distinguish different error classes, models, or code paths without overstating evidence." }, "evaluator_used": true, "retried": false, @@ -615,8 +567,8 @@ ], "soft_pairs": [ "issue:39290|issue:43531", - "issue:43792|issue:45901", "issue:43854|issue:43866", + "issue:43792|issue:45901", "issue:41093|issue:43638", "issue:43425|issue:45070", "issue:43596|issue:45800", @@ -654,278 +606,266 @@ "issue:43645|issue:45399", "issue:43404|issue:45005" ], - "prepared_review_unit_hash": "3f190d492f56b285751f5ceea5cca76ffabba7c6abca9e087bf0596a080676c5", + "prepared_review_unit_hash": "8cbde3b4a90ad0a27117859a33b3783e37858211c7834e571f5c56158a9aae4e", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 39, - "serialized_chars": 39123, - "estimated_input_tokens": 9781, - "estimated_eval_tokens": 19818 + "serialized_chars": 39133, + "estimated_input_tokens": 9784, + "estimated_eval_tokens": 19824 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly unrelated issue reports. The closest match is the ZeRO-3 index-out-of-bounds pair (41093/43638), which shares a similar failure shape, but the packet only supports soft similarity rather than a confirmed duplicate.", - "confidence": 0.78, - "canonical_issue_reason": "issue:41093 is a reasonable canonical representative because it is a concrete, well-described ZeRO-3 index error report and the clearest anchor for the only near-match in the cluster. That said, the evidence for a true duplicate relationship with 43638 is only soft similarity, not a confirmed duplicate.", + "summary": "This cluster is mostly a collection of unrelated Transformers bug reports with a few broad theme overlaps (sliding-window configs, ZeRO-3, dtype mismatches, Qwen/Gemma variants). Only one pair looks like a likely true duplicate: the two SigLIP2 issues reporting bad/garbled outputs. No PRs are present.", + "confidence": 0.83, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:41093 is the best overall issue representative because it is the most specific and central report around the shared ZeRO-3 empty-dimension/index-0 failure pattern, even though the packet does not establish a definitive duplicate pair.", + "best_issue_reason": "issue:43994 is the best representative of the only likely duplicate area: it is broader and more directly describes the user-facing SigLIP2 failure across AutoModel/pipeline.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:39290", "right": "issue:43531", "accept": false, - "reason": "Both mention sliding-window-related model config behavior, but the concrete failures differ: Gemma3 missing an attribute with vLLM vs Qwen3-MoE sliding_window handling." + "reason": "Both mention sliding_window, but they are different models and different failure modes (Gemma3 config field vs Qwen3-MoE behavior)." }, { - "left": "issue:43792", - "right": "issue:45901", + "left": "issue:43854", + "right": "issue:43866", "accept": false, - "reason": "Whisper runtime failure and table-question-answering crashes are unrelated code paths and different error classes." + "reason": "Unrelated model-loading failure vs corrupted checkpoint report; no shared concrete bug path." }, { - "left": "issue:43854", - "right": "issue:43866", + "left": "issue:43792", + "right": "issue:45901", "accept": false, - "reason": "A GLM-4.7 model-loading test failure and a corrupted Ovis2 checkpoint are different underlying problems." + "reason": "Whisper loading/runtime issue is unrelated to table-question-answering crashes." }, { "left": "issue:41093", "right": "issue:43638", "accept": false, - "reason": "These are closely related ZeRO-3 index-0 out-of-bounds reports, but the packet only supports soft similarity and does not establish a confirmed duplicate or the same exact bug." + "reason": "Both involve index/shape errors, but one is a mask mismatch and the other is a ZeRO-3 index-out-of-bounds on a different model path." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch-version incompatibility and a pydantic PretrainedConfig regression are distinct failures with no shared code-path evidence." + "reason": "Torch version incompatibility and a pydantic PretrainedConfig regression are different bugs." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "Both are version-related breakages, but one is a ZeRO-3/Bert index error while the other is a torch/transformers compatibility report." + "reason": "Both mention environment/deepspeed-ish pain, but one is a ZeRO-3 Bert index error and the other is a torch/transformers version incompatibility." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Incorrect detokenization for DeepSeek Coder is unrelated to a table-question-answering task crash." + "reason": "Detokenization bug in DeepSeek code is unrelated to table-question-answering task crashes." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper model runtime failure and pydantic config breakage are not the same bug." + "reason": "Whisper runtime failure is unrelated to the pydantic/config regression." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Both touch Qwen3.5, but one is a chat crash and the other is a streaming inference API failure on a different model size and path." + "reason": "Both are Qwen3.5 chat-related, but one is a crash in transformers chat and the other is a streaming chat/completions API failure; not the same concrete bug." }, { "left": "issue:44292", - "right": "issue:45901", - "accept": false, - "reason": "Qwen-3-8B-NVFP4 runtime error and table QA crash are unrelated." - }, - { - "left": "issue:39290", - "right": "issue:45823", - "accept": false, - "reason": "Gemma3 sliding-window config and Gemma4 device-map forward mismatch are different issues despite both involving Gemma-family models." - }, - { - "left": "issue:45464", "right": "issue:45685", "accept": false, - "reason": "Qwen streaming API failure and an MPS histogram kernel limitation are unrelated." + "reason": "Qwen-3-8B NVFP4 runtime error is unrelated to an MPS histogram kernel missing for MoE." }, { "left": "issue:43404", "right": "issue:43866", "accept": false, - "reason": "A tied-lm_head bug in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." + "reason": "Tied lm_head weights in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Torch dynamo grouped_mm tracing failure and an MPS int histogram limitation are different backend/model problems." + "reason": "Grouped_mm tracing failure in Mixtral is a different code path than the MPS Int histogram issue." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both involve dtype/mixed-precision themes, but the concrete failures are different: grouped_mm tracing vs autocast mismatch on another model." + "reason": "Both are dtype/runtime issues, but they affect different models and different execution paths (dynamo tracing vs autocast)." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Processor failure for Voxtral and a template-node compile 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 Gemma3 sliding-window attribute failure are unrelated." + "reason": "ImportError for GenerationMixin is unrelated to sliding-window config issues." }, { "left": "issue:45800", "right": "issue:45823", "accept": false, - "reason": "Torch/transformers incompatibility and Gemma4 device-map forward mismatch are different bug reports." + "reason": "Version incompatibility is not the same bug as a Gemma4 device mismatch during forward." }, { "left": "issue:41093", "right": "issue:41762", "accept": false, - "reason": "Both are ZeRO-3 index errors, but they involve different models and setup contexts, so the same underlying bug is not established." + "reason": "Both are zero/shape index errors, but one is a mask-length mismatch and the other is a loading-time size-0 index error on Gemma3; too different to treat as the same bug." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are different model-support requests/failures." + "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are separate model-format issues." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is not the same as the pydantic PretrainedConfig break." + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to the pydantic PretrainedConfig regression." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Torch dynamo tracing failure and GPT-OSS MoE TP-only not working are separate issues." + "reason": "Mixtral torch-dynamo tracing failure and GPT-OSS tensor-parallel support are different problems." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "CUDA reserved-memory OOM under int4 quantization is not the same as tied-weights behavior on translation models." + "reason": "Int4 memory/OOM regression is unrelated to tied weights in translation models." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE and Qwen3.5 streaming API failure are unrelated." + "reason": "Phi-tiny-MoE autocast dtype mismatch is unrelated to Qwen3.5 streaming chat failures." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "`init_empty_weights` unexpected argument handling and pydantic config-field breakage are different regressions." + "reason": "Unexpected _is_hf_initialized argument during init_empty_weights is a different bug than the pydantic field break." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF gpt-oss support and Qwen3.5 streaming inference failure do not match." + "reason": "GPT-OSS GGUF support and Qwen3.5 chat streaming are different code paths." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code breakage and torch/transformers incompatibility are too broad and not the same concrete bug." + "reason": "Remote_code breakage and a torch/transformers version incompatibility are not the same underlying bug." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to the pydantic PretrainedConfig regression." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to the pydantic config regression." }, { "left": "issue:43541", "right": "issue:44589", "accept": false, - "reason": "Grouped_mm tracing failure and Float8 storage lookup failure are different backend/runtime errors." + "reason": "Grouped_mm tracing failure and missing Float8 storage support are different runtime issues." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Both are ZeRO-3 index errors, but the model and initialization context differ enough that they do not clearly describe the same bug." + "reason": "Both are index errors under ZeRO-3, but they involve different models and different failure shapes." }, { "left": "issue:43054", "right": "issue:43994", - "accept": false, - "reason": "Siglip2 text-embedding quality regression and a Siglip2 AutoModel/pipeline issue are not the same failure." + "accept": true, + "reason": "Both report SigLIP2 producing bad/garbled outputs; same underlying model quality/implementation problem across different entry points." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite Qwen config unwrapping and uninitialized-memory in rotary embedding are different bugs." + "reason": "Composite Qwen config unwrapping and uninitialized rotary embedding memory are distinct Qwen3.5 issues." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "Tensorboard-without-TensorFlow backend error and torch/transformers version incompatibility are unrelated." + "reason": "TensorBoard-without-TensorFlow backend error is unrelated to the torch/transformers incompatibility." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP weight dimensionality error and TRL SFT embedding unpacking failure are different training-path problems." + "reason": "FSDP 2-D weight error and TRL input-embedding unpacking failure are unrelated." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Mistral-3 image preprocessor dtype mismatch and Qwen3.5 TRL embedding unpacking are unrelated." + "reason": "Mistral3 image-preprocessor dtype mismatch is unrelated to Qwen3.5 embedding unpacking." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "High perplexity on gpt-oss-20b and a generic GLM5 issue are not the same bug." + "reason": "High perplexity on gpt-oss-20b and GLM5 are different model-specific issues." }, { "left": "issue:45632", "right": "issue:45698", "accept": false, - "reason": "Both involve custom/remote code loading, but one is a cache-path collision and the other is wrong custom module loading after save_pretrained; not enough evidence for the same defect." + "reason": "Both involve remote_code/custom-module loading, but one is a leaf-directory cache collision and the other is a wrong-module load after save_pretrained; not clearly the same concrete bug." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Jupyter custom-model initialization breakage and flash-attn2 fallback blocking are separate regressions." + "reason": "Custom models in Jupyter and fallback to flash-attn2 being blocked are unrelated." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Tied weights in a specific multimodal Mistral3 model and tied-weights issues on translation models are related in theme but not the same concrete bug." + "reason": "Specific tied-weight bug in Mistral3 is not the same as the broader translation-model tied-weights report." } ] }, "evaluator_result": { - "accept": false, - "feedback": "The summary is mostly grounded, but it overstates the evidence for one key soft edge: issue:41093 vs issue:43638 is not supported as an accepted duplicate in the packet (the packet only shows soft similarity, with deterministic_accept=false). The representative/canonical choice can still be plausible, but the wording should be more conservative: there is similarity around ZeRO-3 index-out-of-bounds failures, yet the packet does not establish a confirmed duplicate pair. The rest of the rejected soft edges are broadly conservative." + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft pair, issue:43054 vs issue:43994, is reasonably supported by the titles/body summaries as a likely SigLIP2 duplicate area. The other verdicts are appropriately cautious and do not overclaim. No PR pair issues are present." }, "evaluator_used": true, - "retried": true, - "accepted_nontrivial_soft_edge": false, + "retried": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -954,7 +894,6 @@ "issue:43493", "issue:43643", "issue:43644", - "issue:43782", "issue:43828", "issue:43856", "issue:43866", @@ -979,15 +918,16 @@ "issue:44910", "issue:44918", "issue:44928", - "issue:44960", "issue:45005", "issue:45020", "issue:45237", "issue:45399", "issue:45507", "issue:45663", + "issue:45685", "issue:45701", "issue:45800", + "issue:45823", "issue:45859", "issue:45902" ], @@ -1020,251 +960,233 @@ "issue:43054|issue:43493", "issue:44403|issue:44451", "issue:43975|issue:45237", - "issue:43643|issue:45020", "issue:44403|issue:44488", + "issue:43643|issue:45020", "issue:42947|issue:43856", "issue:41762|issue:42915", "issue:43881|issue:43883", - "issue:43782|issue:44960" + "issue:45685|issue:45823" ], - "prepared_review_unit_hash": "10a32933fdd6a8abde44c70c9c8088b5f6b0d2c881da0a0605eb210163a88af5", + "prepared_review_unit_hash": "b813b2861095e130a357b2e6a92620d90dd67c93565871530e803f060013445c", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37508, - "estimated_input_tokens": 9377, - "estimated_eval_tokens": 19010 + "serialized_chars": 37575, + "estimated_input_tokens": 9394, + "estimated_eval_tokens": 19044 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated issue reports across different models and bugs. The only clear near-duplicate pair is the transformers 5.x non-persistent buffer regression: issue 43950 and issue 44534, with 45902 looking like a concrete instance of the same underlying problem.", - "confidence": 0.87, - "canonical_issue_reason": "Issue 43950 is the clearest, broadest description of the regression ('from_pretrained() silently corrupts non-persistent buffers'), making it the best canonical issue for this bug family.", + "summary": "This cluster is mostly heterogeneous. The only clear duplicate family is the Transformers v5 non-persistent-buffer regression; most other pairs are merely same-subsystem or same-model-family similarities, not the same concrete bug.", + "confidence": 0.95, + "canonical_issue_reason": "issue:43950 is the clearest anchor for the shared v5 non-persistent-buffer regression: it states the core bug plainly, is broadly applicable, and matches the accepted duplicate follow-up issues.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43950 is the best tracking issue overall for the non-persistent-buffer bug: it states the root regression clearly and is broader than the later, more specific reports.", + "best_issue_reason": "issue:43950 is the best overall representative because it captures the underlying loader/materialization regression more precisely than the narrower follow-up reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "Both are loading/compatibility failures, but they concern different mechanisms and likely different fixes (Qwen3Moe FP8 config vs flash-attn fallback gating)." + "reason": "Both involve loading/runtime checks, but one is Qwen3 MoE FP8 and the other is flash-attn2 fallback gating; not the same bug." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache crash on Qwen3.5 is a runtime cache bug; tied-weights issues on translation models are a separate model-loading regression." + "reason": "PagedAttentionCache crash on Qwen3.5 is unrelated to tied-weights handling in translation models." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization/offloading behavior is unrelated to tied-weight handling for translation models." + "reason": "Quantized offloading behavior is a different code path from tied-weights/v5 translation-model issues." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "NemotronH checkpoint loading failure is a model-specific loading issue, not the tied-weights regression." + "reason": "NemotronH checkpoint loading failure is not the same as tied-weights regressions." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "DSA indexer missing ReLU and a documentation typo about the removed run command are unrelated." + "reason": "DSA indexer ReLU discussion and a docs typo are unrelated." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "These involve different model paths and failures: Qwen3.5 unpacking with TRL vs GraniteMoEHybrid calling an invalid method." + "reason": "TRL SFT embedding unpacking and GraniteMoEHybrid invalid method are different failures." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A cache-strategy typo and stale v5 docs about the run command are both documentation issues, but not the same change or bug." + "reason": "Cache-strategy typo is documentation-only and unrelated to removed run-command docs." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "DeepSeek detokenization in v5 and Qwen-3 NVFP4 runtime errors are different model-specific problems." + "reason": "Detokenization regression and Qwen-3-8B-NVFP4 runtime error are different concrete bugs." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "No shared code path is evident; one is NVFP4 execution, the other is a GraniteMoE method error." + "reason": "NVFP4 runtime failure and GraniteMoEHybrid invalid method do not share a concrete code path." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Both describe the same transformers 5.x regression where non-persistent buffers are corrupted/junk-filled during loading." + "reason": "Both describe the same v5 non-persistent-buffer regression: buffers are corrupted/junk-filled during loading/materialization." }, { - "left": "issue:43644", + "left": "issue:43643", "right": "issue:43950", "accept": false, - "reason": "Missing fields from trust_remote_code config loading is a separate AutoConfig issue, not buffer corruption." + "reason": "Missing fields in trust_remote_code config output is a different symptom from non-persistent-buffer corruption." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "SigLIP2 docs example failures and SigLIP2 text-embedding quality discrepancies are related only by model name, not by the same bug." + "reason": "SigLIP2 docs example errors are not the same as poorer text embeddings." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Autocast dtype mismatch on Phi-tiny-MoE and Qwen3.5 gradient explosion from 3D position_ids are different failure modes and code paths." + "reason": "Autocast dtype mismatch and TRL SFT embedding unpacking are different failures." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "CustomTrainer compute_loss kwarg incompatibility and Gemma3 ZeRO-3 loading IndexError are unrelated training/loading bugs." + "reason": "Trainer compute_loss signature mismatch and Gemma3 ZeRO-3 loading error are unrelated." }, { "left": "issue:41628", "right": "issue:45800", "accept": false, - "reason": "An AutoImageProcessor import failure is not the same as a torch/transformers version incompatibility report." + "reason": "AutoImageProcessor import failure and torch/transformers version incompatibility are distinct issues." }, { "left": "issue:43950", "right": "issue:45902", "accept": true, - "reason": "45902 is a concrete manifestation of the same non-persistent-buffer corruption problem, with inv_freq left uninitialized after materialization." + "reason": "The Qwen3_5Moe inv_freq uninitialized-memory report is a model-specific manifestation of the same non-persistent-buffer materialization regression." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Bitsandbytes constructor incompatibility and missing Float8 storage support are different compatibility errors." + "reason": "bitsandbytes Int8Params keyword incompatibility and Float8 storage lookup failure are different compatibility bugs." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "A corrupted Ovis2 checkpoint is unrelated to a generic loading-noise report." + "reason": "Corrupted checkpoint and generic loading noise are not the same underlying issue." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality and version-dependent tokenization changes are different issues." + "reason": "SigLIP2 embedding quality and tokenization changes are not the same bug." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to Molmo missing all_tied_weights_keys." + "reason": "Bad AutoTokenizer error message for Voxtral and Molmo missing all_tied_weights_keys are unrelated." }, { "left": "issue:38175", "right": "issue:43054", "accept": false, - "reason": "Both mention SigLIP2, but zero probabilities and worse embeddings are not clearly the same underlying defect." + "reason": "Both are SigLIP2-related, but zero probabilities and worse text embeddings are not clearly the same concrete defect." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Gemma4 FSDP2 shared_kv_states KeyError and Qwen3_5Moe uninitialized inv_freq are distinct model bugs." + "reason": "Gemma4 FSDP2 shared_kv_states rebuilding and Qwen3_5Moe inv_freq memory initialization are unrelated." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "FlashAttention illegal memory access from 3D position_ids is unrelated to CPU offload incompatibility in Qwen3_5MoeTextRotaryEmbedding." + "reason": "Qwen3.5 FlashAttention position_ids crash and CPU-offload incompatibility are different bugs." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Tokenizer loading error messaging and glm-4v-9b loading failure are different problems." + "reason": "Voxtral AutoTokenizer error messaging and glm-4v-9b loading failure are unrelated." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "LoRA gradient-checkpointing behavior and int4 quantization memory growth are separate training/runtime regressions." + "reason": "LoRA gradient checkpointing ineffectiveness and int4 quantization OOM are different memory issues." }, { "left": "issue:43054", "right": "issue:43493", "accept": false, - "reason": "Both are about SigLIP2 behavior, but one is quality/embedding mismatch while the other is HF-vs-JAX implementation discrepancy; not the same concrete bug." + "reason": "Both concern SigLIP2 quality, but one is embedding quality and the other is HF-vs-JAX discrepancy; not enough to treat as the same bug." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Generic loading noise and inability to load a specific model are not the same issue." - }, - { - "left": "issue:43975", - "right": "issue:45237", - "accept": false, - "reason": "DeepSeek detokenization and GPT-OSS-20B on AMD GPUs are unrelated." - }, - { - "left": "issue:43643", - "right": "issue:45020", - "accept": false, - "reason": "Missing fields in AutoConfig with trust_remote_code and broader remote_code breakage are related in theme but not clearly the same underlying bug." + "reason": "Generic loading noise and a specific model load failure are not the same underlying issue." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Two model-loading reports for different checkpoints; no evidence they share the same root cause." - }, - { - "left": "issue:42947", - "right": "issue:43856", - "accept": false, - "reason": "Gradient-checkpointing inefficiency with LoRA and Qwen3 MoE memory usage are both memory-related but not the same code-path failure." + "reason": "Generic loading noise and another model-specific load failure are not the same bug." }, { - "left": "issue:41762", - "right": "issue:42915", + "left": "issue:43643", + "right": "issue:45020", "accept": false, - "reason": "Gemma3 ZeRO-3 loading IndexError and Qwen3Moe FP8 config failure are different loading paths." + "reason": "Remote-code config field loss is related in theme, but too different in concrete failure from broad remote_code breakage." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "glm-4v-9b loading failure and Molmo missing all_tied_weights_keys are unrelated model-specific issues." + "reason": "Both are load failures, but they point to different models and different failure modes." }, { - "left": "issue:43782", - "right": "issue:44960", + "left": "issue:45685", + "right": "issue:45823", "accept": false, - "reason": "Qwen3VL weight-only loading error and GLM5 are not the same bug." + "reason": "MPS histogram dtype error in MoE and Gemma4 device mismatch are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The canonical issue choice (43950) is well supported by the titles, and the two accepted soft-edge pairs are strong matches on the same non-persistent-buffer regression. The rejected pairs are framed cautiously and do not overstate shared causality." + "feedback": "Grounded overall. The summary stays conservative: 43950/44534 are clearly the same non-persistent-buffer regression, and 45902 is a plausible model-specific manifestation of that same bug. The remaining pairs are appropriately rejected as different concrete failures rather than the same issue." }, "evaluator_used": true, "retried": false, @@ -1298,6 +1220,7 @@ "issue:43404", "issue:43638", "issue:43742", + "issue:43782", "issue:43792", "issue:43824", "issue:43827", @@ -1319,23 +1242,21 @@ "issue:44805", "issue:44863", "issue:44928", + "issue:44960", "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:43782|issue:44960", "issue:44991|issue:45399", "issue:43824|issue:43872", "issue:42491|issue:44368", @@ -1345,8 +1266,8 @@ "issue:45081|issue:45399", "issue:45020|issue:45356", "issue:43742|issue:45399", - "issue:43404|issue:43950", "issue:43827|issue:44509", + "issue:43404|issue:43950", "issue:43638|issue:45823", "issue:42915|issue:45020", "issue:43872|issue:44291", @@ -1366,235 +1287,234 @@ "issue:44479|issue:44560", "issue:44530|issue:44863", "issue:24643|issue:30064", - "issue:43366|issue:43828", - "issue:45325|issue:45902" + "issue:43366|issue:43828" ], - "prepared_review_unit_hash": "7ac17176410c5ced5bf7ed2771b07e9ceb19fec2ae927b91cfa357a113e0e113", + "prepared_review_unit_hash": "f23e573b746e9226d625d89d680de8705663d82ecb8cc7bd562bf4ef5dd4743d", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 38072, - "estimated_input_tokens": 9518, - "estimated_eval_tokens": 19292 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 37042, + "estimated_input_tokens": 9261, + "estimated_eval_tokens": 18778 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly overbroad and heterogeneous. I only see a few true duplicate subclusters: v5 docs mentioning removed pipeline tasks, the `_is_hf_initialized` loading TypeError, and the mask-shape mismatch error. Most other pairs are only superficially similar and should stay separate.", - "confidence": 0.66, - "canonical_issue_reason": "No single canonical issue for the whole cluster: the items split into unrelated bug families, with only a few small duplicate subclusters.", + "summary": "The cluster is mostly a set of unrelated regression reports; only two pairs look like true duplicates: the v5 docs/pipeline cleanup and the `_is_hf_initialized` init/loading incompatibility. Everything else is too model- or code-path-specific to merge together.", + "confidence": 0.58, + "canonical_issue_reason": "44509 is the clearest canonical issue for the docs/pipeline-removal bug: it is broader and more explicit than 43827, which is the narrower variant of the same documentation problem.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue across the full cluster. If forced to pick a representative for one subcluster, 44509 is the clearest docs-related issue, but it does not represent the other bugs here.", + "best_issue_reason": "44509 is the best issue to keep as the representative of the only clear duplicate group in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45685", - "right": "issue:45823", + "left": "issue:43782", + "right": "issue:44960", "accept": false, - "reason": "Different model/runtime failures: MPS histogram on Int vs Gemma4 PLE device mismatch." + "reason": "Different models and failure modes; one is Qwen3VL weight-only loading, the other is a GLM5 model request." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer loading regression and flash-attn fallback gating are unrelated code paths." + "reason": "Tokenizer loading failure vs flash-attn fallback gating; different bugs." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Importing a missing Qwen2.5-VL class is unrelated to the bitsandbytes `_is_hf_initialized` TypeError." + "reason": "ImportError for a missing class vs bitsandbytes Int8Params kwarg incompatibility." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "Different problems: LoRA checkpoint compatibility across versions vs a tied-embeddings warning during Qwen3.5 fine-tuning." + "reason": "LoRA/Qwen3 MoE compatibility problem vs a gradient-checkpointing warning; not the same bug." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Broad remote_code regression is not the same bug as a Gemma4 device-map forward mismatch." + "reason": "Broad remote_code breakage vs a Gemma4 device_map forward mismatch." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Tokenizer-mapping registry failure is a different path from the remote_code loading breakage." + "reason": "Tokenizer mapping/model registration issue vs remote_code regressions; different code paths." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Autocast dtype mismatch and device placement mismatch are different failures." + "reason": "Autocast dtype mismatch vs device placement mismatch; not the same failure." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Mistral regex patch crash is unrelated to flash-attn2 fallback blocking." + "reason": "Tokenizer regex patch crash vs flash-attn fallback logic; unrelated." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "Remote_code loading regressions and Kimi codec handling warnings are different underlying bugs." + "reason": "General remote_code breakage vs a specific Kimi-K2.5 tokenizer/codec regression." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "MobileLLM key error is unrelated to flash-attn2 installation/fallback checks." + "reason": "KeyError while loading a model vs flash-attn fallback checks; unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Same v5 docs cleanup for removed summarization/translation pipeline tasks." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Tied lm_head weights and non-persistent buffer corruption are separate model-loading bugs." + "reason": "Tied lm_head bug vs non-persistent buffer corruption; different issues." }, { "left": "issue:43638", "right": "issue:45823", "accept": false, - "reason": "DeepSpeed ZeRO-3 index error with BERT is unrelated to Gemma4's device mismatch." - }, - { - "left": "issue:43827", - "right": "issue:44509", - "accept": true, - "reason": "Both report the same docs drift: v5 docs still reference removed pipeline tasks." + "reason": "Zero-3 index error on a Bert model vs Gemma4 PLE device mismatch." }, { "left": "issue:42915", - "right": "issue:45923", + "right": "issue:45020", "accept": false, - "reason": "FineGrainedFP8Config failure and missing supports_gradient_checkpointing are different issues." + "reason": "FP8/Qwen3 MoE failure vs broad remote_code regressions; not the same bug." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same underlying `_is_hf_initialized` constructor/loading regression causing TypeError during model load." + "reason": "Both report the same `_is_hf_initialized` incompatibility during model initialization/loading." }, { "left": "issue:43257", "right": "issue:43866", "accept": false, - "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed is unrelated to a corrupted Ovis2 checkpoint." + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed vs a corrupted Ovis2 checkpoint." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Whisper runtime failure and `_torch_extract_fbank_features()` are not the same concrete bug." + "reason": "Whisper runtime failure vs a generic fbank feature extraction issue; too indirect to treat as duplicate." }, { "left": "issue:42915", "right": "issue:43872", "accept": false, - "reason": "Qwen3 FP8 config failure and `_is_hf_initialized` loading error are different failure modes." + "reason": "Different root causes: FP8 config failure vs bitsandbytes/init kwarg mismatch." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Bad Voxtral tokenizer error messaging is unrelated to EMBEDDIA/est-roberta tokenizer loading failure." + "reason": "Bad AutoTokenizer error message for Voxtral vs tokenizer loading failure for a specific model." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Trainer signature mismatch and mask shape mismatch are completely different bugs." + "reason": "Unexpected compute_loss kwarg vs tensor mask shape mismatch; unrelated." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "ZeRO-3 weight-loading failure and gradient averaging with GAS are separate DeepSpeed issues." + "reason": "Weights not converted in Qwen3 MoE loading vs gradient averaging under DeepSpeed." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing ineffective with PEFT LoRA is not the same as a missing trust_remote_code flag." + "reason": "LoRA gradient checkpointing ineffective vs missing supports_gradient_checkpointing flag; related theme but different bug." }, { "left": "issue:43940", "right": "issue:45313", "accept": false, - "reason": "Both involve ZeRO-3 loading, but these are different Qwen variants with different missing-weight symptoms." + "reason": "Different Qwen variants and different ZeRO-3 load failures; not clearly the same concrete bug." }, { "left": "issue:43854", "right": "issue:44863", "accept": false, - "reason": "GLM-4.7-Flash unit-test load failure and NemotronH checkpoint loading are unrelated model-specific bugs." + "reason": "Model-loading failure in unit tests vs NemotronH checkpoint-loading failure; different models." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "RLHF NaN/BF16 collapse from 3D position_ids is unrelated to the flash_attention s_aux=None crash." + "reason": "RLHF gradient explosion/NaN issue vs flash_attention.py crash on s_aux=None." }, { "left": "issue:43872", "right": "issue:45538", "accept": false, - "reason": "Bitsandbytes `_is_hf_initialized` loading issue is unrelated to CLIPTokenizer's model_max_length constant." + "reason": "bitsandbytes init kwarg incompatibility vs CLIPTokenizer max length bug." }, { "left": "issue:41093", "right": "issue:44805", - "accept": true, - "reason": "Same mask/index shape-mismatch error pattern; these look like the same underlying bug." + "accept": false, + "reason": "Both are IndexErrors, but the mask mismatch and the generic shape mismatch are different bugs." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Voxtral loading error-message issue and sleng-bert loading failure are different model-specific problems." + "reason": "Voxtral tokenizer error messaging vs a specific model loading failure." }, { "left": "issue:44479", "right": "issue:44560", "accept": false, - "reason": "The former is a broad video-input regression across Qwen VL/MoE models; the latter is a specific StopIteration in Qwen3-vl-embedding." + "reason": "Related video-input regressions, but the impacted models and failure details are different enough to keep separate." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3.5 PagedAttention crash vs NemotronH checkpoint loading; unrelated." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed training weight-shape error and void segmentation map processing are unrelated." + "reason": "DeepSpeed training weight-dimension error vs void segmentation-map preprocessing; unrelated." }, { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "GGUF gpt-oss support and an autocast dtype mismatch are not the same code-path problem." - }, - { - "left": "issue:45325", - "right": "issue:45902", - "accept": false, - "reason": "Qwen2.5-VL rope-index scaling and Qwen3_5Moe uninitialized-memory reads are different rotary-embedding bugs." + "reason": "GGUF architecture support request vs autocast dtype mismatch; not the same issue." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analyst output is grounded in the packet and stays conservative. The three accepted soft pairs are plausible same-bug/duplicate subclusters, and the rejections are appropriately cautious given the titles and descriptions. The overall summary correctly treats the cluster as heterogeneous rather than forcing a single canonical issue." + "feedback": "The verdicts are grounded and conservative. The two accepted pairs are well supported by the packet: 43827/44509 are the same v5 docs pipeline-removal cleanup, and 43872/44291 are the same `_is_hf_initialized` loading incompatibility. The rejections generally stay within the evidence and avoid overclaiming duplicates." }, "evaluator_used": true, "retried": false, @@ -1632,13 +1552,11 @@ "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", @@ -1653,21 +1571,23 @@ "issue:44918", "issue:44960", "issue:44977", - "issue:44991", "issue:45030", "issue:45276", "issue:45313", + "issue:45325", "issue:45357", "issue:45584", "issue:45663", "issue:45685", "issue:45823", - "issue:45834" + "issue:45834", + "issue:45902" ], "soft_pairs": [ + "issue:45325|issue:45902", "issue:45663|issue:45823", - "issue:44155|issue:45584", "issue:43366|issue:43531", + "issue:44155|issue:45584", "issue:43950|issue:45834", "issue:44560|issue:45685", "issue:43824|issue:43957", @@ -1695,249 +1615,228 @@ "issue:44451|issue:44488", "issue:42915|issue:44756", "issue:45313|issue:45357", - "issue:42491|issue:45276", - "issue:43708|issue:44075", - "issue:44488|issue:44991", - "issue:43644|issue:43645" + "issue:42491|issue:45276" ], - "prepared_review_unit_hash": "aa108acc7cf91cbc43d5ded04c4d24872ffc4ba184dde8913e9a4591a9af742b", + "prepared_review_unit_hash": "11bb5cf7a432dd1d7b343b408584ac953985b0311e3a3642602b27199accccc8", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 34, - "serialized_chars": 37789, - "estimated_input_tokens": 9448, - "estimated_eval_tokens": 19152 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36682, + "estimated_input_tokens": 9171, + "estimated_eval_tokens": 18598 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is mostly heterogeneous: several unrelated Transformers regressions are mixed together, but there are a few true duplicate subclusters around non-persistent buffers, custom model initialization, and SigLIP2 output corruption.", - "confidence": 0.73, - "canonical_issue_reason": "No single canonical issue fits the whole cluster; 43644 is the best local canonical for the non-persistent-buffer regression because it states the shared bug clearly and has the closest matching duplicate wording.", + "summary": "This is a loose mixed cluster: a few genuine duplicates around Transformers v5 regressions, but most items are unrelated model-specific bugs. I only accepted the clearly same-root-cause pairs.", + "confidence": 0.62, + "canonical_issue_reason": "43646 is the broadest, clearest regression report in the set: it captures the core Transformers v5 custom-model initialization breakage that best represents the small duplicate-heavy subset.", "canonical_pr_reason": null, - "best_issue_reason": "43644 is the strongest overall issue candidate: it is concrete, reproducible, and anchors a real duplicate subcluster better than the more model-specific or orthogonal reports.", + "best_issue_reason": "It is the most representative issue for the cluster\u2019s regression-heavy subset and is cleaner than the highly model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45663", - "right": "issue:45823", + "left": "issue:45325", + "right": "issue:45902", "accept": false, - "reason": "Different Gemma4 failures: an FSDP2 shared_kv_states KeyError vs a device_map='auto' forward mismatch. Same model family, but not the same bug." + "reason": "Different bugs and different model families: Qwen2.5-VL rope scaling vs Qwen3.5-MoE inv_freq materialization." }, { - "left": "issue:44155", - "right": "issue:45584", + "left": "issue:45663", + "right": "issue:45823", "accept": false, - "reason": "Unrelated symptoms and code paths: AudioFlamingo batched-inference leakage vs Whisper empty-transcription handling." + "reason": "Both are Gemma4, but one is an FSDP2 shared_kv_states KeyError and the other is a device_map forward mismatch; not the same code-path." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "One is GGUF/gpt-oss architecture support; the other is a Qwen3-MoE sliding_window issue. Same general area, different change." + "reason": "GGUF GPT-OSS support and Qwen3-MoE sliding_window are unrelated changes." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "AudioFlamingo3 track-leak batching bug vs Whisper empty-transcription failure; different models and failures." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "Non-persistent-buffer corruption vs Kosmos2.5 long-OCR index error are different bugs and different model paths." + "reason": "Non-persistent buffer corruption is unrelated to Kosmos2.5 long-OCR index errors." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Qwen3-vl video StopIteration and an MoE MPS histogram implementation error are unrelated." + "reason": "Qwen3-VL video StopIteration and MoE MPS histogram dtype error are unrelated." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Import error for a missing VL class vs meta-device loading regressions; too broad and not the same concrete bug." + "reason": "Missing Qwen2_5_VL import and generic meta-device loading breakage are different issues." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Bamba torch.compile/SDPA failure and Qwen3.5 flash-attention illegal memory access are different model/runtime bugs." + "reason": "Bamba torch.compile+SDPA crash is unrelated to Qwen3.5 flash-attention 3D position_ids handling." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "These are effectively the same custom-model initialization regression: one mentions Jupyter specifically, the other is the general form of the same failure." + "reason": "Both report the same Transformers v5 custom model initialization regression; the notebook wording in 43645 is just a context variant." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Qwen3Moe FP8 config failure and meta-device loading breakage do not share the same underlying code path." + "reason": "Qwen3-MoE FP8 failure and meta-device loading failures are different bugs." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "GGUF gpt-oss support and Float8 storage lookup failure are different loader/runtime issues." + "reason": "GGUF GPT-OSS support is unrelated to the Float8 storage lookup error." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "3d_parallel.py runtime failure vs GGUF model architecture support are unrelated." + "reason": "3d_parallel.py runtime failure and GGUF GPT-OSS support are unrelated." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Qwen3-vl video StopIteration and TRL embedding unpacking are different failures." + "reason": "Different Qwen3.5/Qwen3-VL audio-video embedding issues; not the same underlying bug." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "GenerationMixin import error and Qwen3Moe FP8 failure are unrelated." + "reason": "GenerationMixin import error and Qwen3-MoE FP8 failure are unrelated." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Tokenizer mapping failure when adding a model vs Gemma4 PLE device mismatch are different issues." + "reason": "Tokenizer-mapping-related add-new-model-like failure is unrelated to Gemma4 device mismatch." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Video embedding StopIteration and mask-shape IndexError are not the same bug." + "reason": "Qwen3-VL video StopIteration and generic mask-shape IndexError are unrelated." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "_is_hf_initialized flag regression and Kosmos2.5 long OCR index error are unrelated." + "reason": "_is_hf_initialized flag regression and Kosmos2.5 OCR index error are different issues." }, { "left": "issue:38175", "right": "issue:43994", "accept": true, - "reason": "Same SigLIP2 symptom family on the same model: zero probabilities vs nonsensical outputs with AutoModel/pipeline point to the same underlying regression." + "reason": "Both are reports of the same SigLIP2 base model producing broken/zeroed outputs through normal loading/inference paths." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "DeepSpeed ZeRO-3 initialization on a non-pretrained Bert model is a different path from add-new-model-like/tokenizer mapping handling." + "reason": "Zero3 BERT index error and tokenizer-mapping add-new-model-like failure are different code paths." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "ZeRO-3 BertModel index error and TOKENIZER_MAPPING_NAMES handling are different bugs." + "reason": "Zero3 BertModel initialization failure and tokenizer-mapping failure are unrelated." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "Tokenizer mapping issue and mask/tensor shape mismatch are unrelated." + "reason": "Tokenizer mapping validation and mask/tensor shape mismatch are unrelated." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Strict config rejection for granite_speech and tiny-random glm4v config validation are different model/config changes." + "reason": "Both mention stricter config handling, but the model families and failure modes differ; not enough evidence of one underlying bug." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "device_map='auto' dequantized-model offload bug and Strix Halo mmap OOM are both loading-related but not the same concrete failure." + "reason": "gpu+cpu offload dequantized loading failure and Strix Halo mmap OOM are different loading/memory issues." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Both describe the same transformers 5 regression where non-persistent buffers get corrupted/junk values after loading." + "reason": "Same non-persistent-buffer junk regression in Transformers v5; 44534 is a later duplicate of 43644." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "SigLIP2 bad outputs vs documentation example errors/processor mismatch are not the same underlying bug." + "reason": "Same SigLIP2 family, but one is zero probabilities while the other is a doc example/model-processor/quantization error; not the same concrete bug." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "LoRA qwen3_moe hf4.x vs hf5.x compatibility and BitNet packed-weight unpacking are different regressions." + "reason": "Qwen3 MoE LoRA compatibility regression and BitNet packed-weight loading bug are unrelated." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Qwen3 auto device mapping cudaErrorAssert and Qwen3Moe FP8 config failure are different runtime paths." + "reason": "Qwen3 auto device-mapping CUDA assert and Qwen3-MoE FP8 failure are distinct failures." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "GLM5 issue and Qwen3.5 flash-attention generation failure are unrelated." + "reason": "GLM5 and Qwen3.5 flash-attention issues are unrelated." }, { "left": "issue:44451", "right": "issue:44488", "accept": false, - "reason": "Both are model-loading failures, but the titles only show different repos and no clear shared root cause." - }, - { - "left": "issue:42915", - "right": "issue:44756", - "accept": false, - "reason": "Qwen3Moe FP8 failure and mmap OOM on Strix Halo are unrelated." + "reason": "Both are BERT loading failures, but they reference different models and lack evidence of one shared root cause." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "ZeRO-3 language_model weight loading and incorrect visual encoder keys on save_pretrained are different bugs." + "reason": "ZeRO-3 language_model loading and save_pretrained visual-encoder key serialization are different bugs." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "LoRA qwen3_moe checkpoint incompatibility and Gemma4 resize_token_embeddings not updating tied structures are different issues." - }, - { - "left": "issue:43708", - "right": "issue:44075", - "accept": false, - "reason": "resume_from_checkpoint max_steps calculation and SGD args being ignored are unrelated training bugs." - }, - { - "left": "issue:44488", - "right": "issue:44991", - "accept": false, - "reason": "Loading a specific BERT model and tokenizer loading for EMBEDDIA/est-roberta are similar symptoms but not clearly the same bug." - }, - { - "left": "issue:43644", - "right": "issue:43645", - "accept": false, - "reason": "Non-persistent-buffer corruption and custom model initialization in notebooks are different regressions, even though both are Transformers 5 related." + "reason": "Qwen3 MoE LoRA incompatibility and Gemma4 resize_token_embeddings behavior are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: it identifies a heterogeneous cluster and only calls out a few plausible duplicate subclusters. The accepted soft-edge pairs are conservative and well-supported by the titles, especially 43644/44534 and 43645/43646." + "feedback": "The summary is grounded in the packet, and the accepted soft pairs are conservative. The duplicate calls around 43644/44534 and 43645/43646 are well supported, and the broader cluster description stays appropriately cautious about the many unrelated model-specific issues." }, "evaluator_used": true, "retried": false, @@ -1970,12 +1869,16 @@ "issue:43404", "issue:43425", "issue:43454", + "issue:43644", + "issue:43645", + "issue:43708", "issue:43854", "issue:43856", "issue:43873", "issue:43931", "issue:43957", "issue:43994", + "issue:44075", "issue:44155", "issue:44164", "issue:44291", @@ -1986,9 +1889,8 @@ "issue:44756", "issue:44910", "issue:44912", - "issue:44928", "issue:44960", - "issue:45216", + "issue:44991", "issue:45276", "issue:45292", "issue:45335", @@ -1997,8 +1899,6 @@ "issue:45488", "issue:45588", "issue:45663", - "issue:45684", - "issue:45698", "issue:45704", "issue:45759", "issue:45812", @@ -2007,6 +1907,9 @@ "issue:46050" ], "soft_pairs": [ + "issue:43708|issue:44075", + "issue:43644|issue:43645", + "issue:44488|issue:44991", "issue:43854|issue:43931", "issue:45488|issue:45812", "issue:39692|issue:43994", @@ -2025,268 +1928,265 @@ "issue:41720|issue:44155", "issue:43425|issue:43957", "issue:39692|issue:44960", - "issue:42947|issue:45704", "issue:44402|issue:44488", + "issue:42947|issue:45704", "issue:45663|issue:45759", "issue:43873|issue:44387", "issue:44291|issue:44756", "issue:45663|issue:45859", "issue:44155|issue:45823", - "issue:44387|issue:44756", - "issue:44292|issue:44912", "issue:43335|issue:45663", - "issue:43957|issue:44756", + "issue:44292|issue:44912", + "issue:44387|issue:44756", + "issue:43957|issue:44756", "issue:42947|issue:43381", "issue:45276|issue:45292", - "issue:44164|issue:44756", - "issue:43854|issue:45216", - "issue:43381|issue:44928", - "issue:45684|issue:45698" + "issue:44164|issue:44756" ], - "prepared_review_unit_hash": "ee630a243e6efde6c86f10a44b6a2a2b7c41fd9cb5e6973755b7a3d6c99a846f", + "prepared_review_unit_hash": "638c884d5c32dd2d49d75e97c16b9d2d2c0c629d485f0a6671604064161cf1cc", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 37799, - "estimated_input_tokens": 9450, - "estimated_eval_tokens": 19156 + "serialized_chars": 38298, + "estimated_input_tokens": 9575, + "estimated_eval_tokens": 19406 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly heterogeneous cluster of unrelated model-loading, tokenizer, quantization, and memory issues. The only clear duplicate pair is the resize_token_embeddings regression in Gemma4/t5gemma; most other soft links are too broad or model-specific to merge safely.", - "confidence": 0.79, - "canonical_issue_reason": "issue:45292 is the best canonical issue because it captures the generic resize_token_embeddings/output_embeddings regression, while issue:45276 looks like a model-specific Gemma4 manifestation of the same bug.", + "summary": "Most items are separate Transformers regressions across different models and subsystems. Only the resize_token_embeddings reports for Gemma4/generic output embeddings look like a plausible duplicate pair.", + "confidence": 0.67, + "canonical_issue_reason": "issue:45292 is the most generic description of the shared resize_token_embeddings regression; issue:45276 reads like a Gemma4-specific manifestation with extra affected modules.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 is the strongest representative of the only clear duplicate pair and is phrased most generally, making it the best issue to anchor the cluster.", + "best_issue_reason": "issue:45292 is the cleanest representative of the only likely duplicate pair and has the broadest, most reusable title.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Different areas: checkpoint resume step math vs SGD optimizer args being ignored." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "One is non-persistent buffer initialization junk; the other is Jupyter custom-model initialization, which is unrelated." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading failures, but for different models and likely different root causes." + }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Both are load failures, but they involve different models and different symptoms (unit-test load issue vs weight-shape mismatch), so they are not the same bug." + "reason": "Different models and symptoms: GLM-4.7 unit-test load issue vs Qwen3-VL weight-shape mismatch." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but the mechanisms differ: ByteLevel pre-tokenizer being overridden vs wrong token IDs for Granite models." + "reason": "Both are tokenizer regressions, but they affect different model families and are not the same concrete bug." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Same SigLIP2 family, but one is a documentation/example error report and the other is a runtime inference/output regression; not one underlying bug." + "reason": "Same model family, but one is a doc-example/quantization mismatch and the other is bad inference results in AutoModel/pipeline." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Completely different model families and failure modes." + "reason": "Different failures: Qwen3 MoE weight conversion under accelerate/deepspeed vs GLM-4.7 loading in tests." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Both mention memory/quantization-adjacent problems, but one is Qwen3 MoE training memory inefficiency and the other is offloading failure under quantization." + "reason": "Both touch quantization/offloading, but the reported failure modes and likely code paths differ." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "Different models and different issues; one is a Qwen3 MoE LoRA compatibility problem, the other is a Qwen3-VL load shape mismatch." + "reason": "Unrelated issues: Qwen3 MoE LoRA compatibility on hf5.x vs Qwen3-VL load shape mismatch." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both concern memory pressure, but the underlying bugs differ (training inefficiency vs int4 quantization reserved-memory regression)." + "reason": "Both mention memory/OOM under quantization, but they are distinct bugs with different symptoms and causes." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different model paths and different crashes; these are not the same code-path problem." + "reason": "Different model-specific crashes: Qwen3.5 chat vs Gemma4Processor missing _tokenizer in serve." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "Different models and different save/load failures; one is GLM5-related, the other is sinq quantization persistence." + "reason": "GLM5 loading vs sinq save/load corruption are separate issues." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "No shared bug: Qwen3 MoE LoRA branch-compatibility issue vs a generic resize_token_embeddings regression." + "reason": "LoRA/Qwen3 MoE compatibility is unrelated to the resize_token_embeddings regression." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "Both are save/load-related, but one is extra_state handling and the other is init_empty_weights/_is_hf_initialized; different failure modes." + "reason": "Different load/save failures: extra_state handling vs init_empty_weights/_is_hf_initialized argument mismatch." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Both touch flash-attention code, but one is a Qwen3.5 position_ids packing bug and the other is an s_aux=None AttributeError in sink-less models." + "reason": "Both are flash_attention crashes, but one is a packed-sequence misinterpretation on Qwen3.5 and the other is s_aux=None handling." }, { "left": "issue:45276", "right": "issue:45335", - "accept": true, - "reason": "Same underlying regression: resize_token_embeddings fails to propagate to model-specific embedding/output layers in two composite model variants." + "accept": false, + "reason": "Similar wording, but different model families and embedding surfaces; not clearly the same concrete bug." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "Unrelated issues: an ImportError for GenerationMixin versus Whisper pipeline return_language behavior." + "reason": "ImportError for GenerationMixin is unrelated to Whisper pipeline return_language behavior." }, { "left": "issue:43404", "right": "issue:43454", "accept": false, - "reason": "Both are about lm_head tying, but they appear to be separate model-specific implementations rather than one shared concrete bug." + "reason": "Both are lm_head tying problems, but they are model-specific reports rather than the same concrete fix." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Different subsystems and symptoms: auto device mapping CUDA assert vs batched inference token/embedding leak." + "reason": "Different problems: Qwen3 auto device-map crash vs AudioFlamingo3 batched inference leakage." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Different problems: Torch version incompatibility vs meta-device loading breakage." + "reason": "Torch version incompatibility is a different issue from meta-device loading failures." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "Different model families and different bugs; docs/example issues are not the same as GLM5 loading." + "reason": "Unrelated model families and failure modes: SigLIP2 docs/inference vs GLM5 loading." }, { - "left": "issue:42947", - "right": "issue:45704", + "left": "issue:44402", + "right": "issue:44488", "accept": false, - "reason": "Different concerns: gradient checkpointing with PEFT LoRA vs T5 using apex.FusedRMSNorm causing a memory leak." + "reason": "Both involve tokenizer/model loading, but the specific failures and affected models are different." }, { - "left": "issue:44402", - "right": "issue:44488", + "left": "issue:42947", + "right": "issue:45704", "accept": false, - "reason": "Both are model-loading complaints, but there is not enough evidence they share the same root cause; likely separate tokenizer/model metadata issues." + "reason": "Gradient checkpointing with PEFT LoRA is unrelated to the apex.FusedRMSNorm memory leak." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Different architectures and failure points: Gemma4 FSDP2 shared_kv_states KeyError vs Qwen multimodal config unwrapping." + "reason": "Different composite-model bugs: FSDP2 shared_kv_states handling vs text_config unwrapping in from_config." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Both are quantization/memory themed, but one is offloading not working and the other is reserved-memory growth under int4; not the same bug." + "reason": "Both are memory issues under quantization, but one is offloading behavior and the other is reserved-memory growth/OOM." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "Unrelated: unexpected _is_hf_initialized argument during init_empty_weights vs mmap causing OOM on Strix Halo." + "reason": "TypeError from init_empty_weights is unrelated to mmap/OOM on Strix Halo." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Different code paths and symptoms: Gemma4 FSDP2 attention state rebuild vs Qwen3.5 rotary embedding not offload-compatible." + "reason": "Different code paths: Gemma4 FSDP2 attention state vs Qwen3 MoE rotary embedding CPU-offload compatibility." }, { "left": "issue:44155", "right": "issue:45823", "accept": false, - "reason": "Different model classes and failures; track leakage in AudioFlamingo3 is not the same as a Gemma4 device-mismatch forward bug." + "reason": "AudioFlamingo3 batch leakage and Gemma4 device-mapping mismatch are unrelated." }, { - "left": "issue:44387", - "right": "issue:44756", + "left": "issue:43335", + "right": "issue:45663", "accept": false, - "reason": "Both are memory-related, but one is an int4 quantization regression and the other is a platform-specific mmap OOM workaround." + "reason": "Different model/config bugs: sparse-layer creation in SwitchTransformers vs Gemma4 FSDP2 attention state." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization formats and failures: NVFP4 run error vs MXFP4 loading/fallback to bf16." + "reason": "Both are quantization-related load failures, but they are for different models and likely different causes." }, { - "left": "issue:43335", - "right": "issue:45663", + "left": "issue:44387", + "right": "issue:44756", "accept": false, - "reason": "Separate model bugs: SwitchTransformers sparse-layer construction versus Gemma4 FSDP2 attention state handling." + "reason": "OOM under int4 quantization is not the same as disabling mmap for Strix Halo memory pressure." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device loading breakage and Strix Halo mmap OOM are different issues." + "reason": "Meta-device loading regressions are distinct from the Strix Halo mmap/OOM workaround." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Both mention gradient checkpointing, but one is ineffective checkpointing under LoRA and the other is an eval-mode limitation." + "reason": "Both mention gradient checkpointing, but one is PEFT LoRA ineffectiveness and the other is eval-mode incompatibility." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same resize_token_embeddings regression; 45292 is the generic form and 45276 is a model-specific instance." + "reason": "Same underlying bug: resize_token_embeddings does not propagate to output embeddings; 45276 is just Gemma4-specific." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "Different areas: save/from_pretrained extra_state handling vs mmap-related OOM." - }, - { - "left": "issue:43854", - "right": "issue:45216", - "accept": false, - "reason": "Different models and different save/load failures; no shared concrete bug is evident." - }, - { - "left": "issue:43381", - "right": "issue:44928", - "accept": false, - "reason": "Unrelated gradient-checkpointing limitation vs Qwen3.5 3D position_ids/SDPA fallback instability." - }, - { - "left": "issue:45684", - "right": "issue:45698", - "accept": false, - "reason": "Both involve custom model save/load, but one is file permission propagation and the other is wrong module resolution after save_pretrained; not the same bug." + "reason": "Save/from_pretrained extra_state handling is unrelated to mmap-induced OOM on Strix Halo." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster summary is grounded in the packet, and the one accepted duplicate family (resize_token_embeddings across 45276/45335/45292) is supported by the issue titles. The remaining soft-edge rejections are conservative and appropriately avoid overclaiming shared root causes." + "feedback": "The summary stays grounded in the packet and the only accepted soft-edge pair is conservative: 45276 and 45292 both describe the same resize_token_embeddings/output_embeddings regression, with 45276 adding Gemma4-specific detail. The rest of the verdicts appropriately avoid overclaiming duplicates across distinct model families or failure modes." }, "evaluator_used": true, "retried": false, @@ -2307,21 +2207,20 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ - "issue:32090", "issue:36032", "issue:39404", "issue:43257", "issue:43299", + "issue:43381", "issue:43425", "issue:43452", - "issue:43596", "issue:43611", "issue:43645", "issue:43646", "issue:43708", "issue:43716", "issue:43828", - "issue:43856", + "issue:43854", "issue:43901", "issue:43906", "issue:43931", @@ -2336,7 +2235,6 @@ "issue:44509", "issue:44530", "issue:44589", - "issue:44671", "issue:44683", "issue:44756", "issue:44805", @@ -2345,30 +2243,35 @@ "issue:44898", "issue:44912", "issue:44918", - "issue:44991", - "issue:45081", + "issue:44928", "issue:45084", + "issue:45216", "issue:45292", "issue:45335", "issue:45588", "issue:45663", + "issue:45684", + "issue:45698", "issue:45701", "issue:45759", "issue:45823" ], "soft_pairs": [ + "issue:43854|issue:45216", + "issue:43381|issue:44928", + "issue:45684|issue:45698", "issue:44479|issue:44877", "issue:45588|issue:45663", "issue:43646|issue:43708", "issue:43645|issue:44164", "issue:43828|issue:43957", "issue:45292|issue:45335", - "issue:44402|issue:44451", "issue:43425|issue:43828", + "issue:44402|issue:44451", "issue:43257|issue:44863", "issue:44589|issue:44756", - "issue:44683|issue:45823", "issue:39404|issue:45701", + "issue:44683|issue:45823", "issue:43425|issue:45823", "issue:43611|issue:43645", "issue:43425|issue:43716", @@ -2377,262 +2280,231 @@ "issue:39404|issue:44912", "issue:44265|issue:44898", "issue:44479|issue:44805", - "issue:43994|issue:44079", "issue:43901|issue:44509", + "issue:43994|issue:44079", "issue:43425|issue:44918", "issue:44589|issue:45084", "issue:43299|issue:43931", "issue:44164|issue:45759", - "issue:36032|issue:43452", - "issue:43596|issue:44683", - "issue:44671|issue:44991", - "issue:39404|issue:44671", - "issue:43856|issue:44756", - "issue:44991|issue:45081", - "issue:44164|issue:44479", - "issue:32090|issue:39404" + "issue:36032|issue:43452" ], - "prepared_review_unit_hash": "934a3a09071aa05f18f17d133ed304cb3119447af249f73f6bd4c4cf8bcef414", + "prepared_review_unit_hash": "22d601de719e104dac5564e55a56ad9acd71aaac6a84a0098ac2a6d7cc4f80a0", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 34, - "serialized_chars": 38059, - "estimated_input_tokens": 9515, - "estimated_eval_tokens": 19286 + "soft_pair_count": 30, + "serialized_chars": 36816, + "estimated_input_tokens": 9204, + "estimated_eval_tokens": 18664 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: many unrelated Transformers v5 regressions mixed with tokenizer, loading, training, and docs issues. I only see one plausible duplicate pair: resize_token_embeddings behavior for generic models vs t5gemma-specific embeddings.", - "confidence": 0.82, - "canonical_issue_reason": "No single true canonical duplicate stands out. If one issue must anchor the cluster, issue 45292 is the best representative because 45335 appears to be the same resize_token_embeddings bug expressed for t5gemma.", + "summary": "Most of the cluster is heterogeneous. The resize_token_embeddings pair is a solid duplicate, but the Qwen3-VL MoE/loading pair is only superficially similar and should not be merged as the same bug.", + "confidence": 0.8, + "canonical_issue_reason": "issue:45292 is the broadest, most reusable statement of the resize_token_embeddings bug; issue:45335 is the t5gemma-specific manifestation of the same missing propagation.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45292 is the strongest global representative: it states the regression in the broadest form, and 45335 is a clear model-specific restatement of the same problem.", + "best_issue_reason": "issue:45292 is the best representative issue in this set because it captures a concrete, generalizable bug pattern with a clear duplicate counterpart.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Both are model-loading regressions, but they involve different models and different failure modes." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Unrelated bugs: gradient checkpointing in eval mode versus Qwen3.5 RLHF NaNs from position_ids/SDPA behavior." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both touch save/load workflows, but one is about file permissions and the other about loading the wrong custom module." + }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Different regressions: video-input handling for Qwen VL models vs Granite speech config loading/strict config." + "reason": "Different model families and different regressions; only broad loading/config similarity." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Unrelated code paths: flash_attention sink-less crash vs Gemma4 FSDP2 shared_kv_states rebuild issue." + "reason": "Different code paths: flash_attention sink handling versus Gemma4 FSDP2 shared_kv_states reconstruction." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Both are v5 issues, but one is custom model initialization while the other is resume_from_checkpoint max_steps calculation." + "reason": "Custom model initialization in notebooks is unrelated to Trainer resume_from_checkpoint max_steps math." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Notebook custom-model initialization regression is unrelated to save/from_pretrained handling of extra_state." + "reason": "Both involve custom-model workflows, but notebook initialization and extra_state save/load handling are different bugs." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Different failures: autocast dtype mismatch for Phi-tiny-MoE vs meta-device loading breakage for other models." + "reason": "Different causes and symptoms: autocast dtype mismatch for Phi-tiny-MoE versus meta-device loading regressions." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Same underlying bug: resize_token_embeddings does not propagate to tied output/decoder embeddings; 45335 is a t5gemma-specific instance of the same regression." + "reason": "Same underlying resize_token_embeddings bug: 45335 is the t5gemma-specific instance of the output embedding not being updated." }, { - "left": "issue:44402", - "right": "issue:44451", + "left": "issue:43425", + "right": "issue:43828", "accept": false, - "reason": "Both involve model/tokenizer loading, but the concrete problems are different and model-specific." + "reason": "Torch version compatibility and a model-specific autocast dtype mismatch are not the same bug." }, { - "left": "issue:43425", - "right": "issue:43828", + "left": "issue:44402", + "right": "issue:44451", "accept": false, - "reason": "Torch version incompatibility is not the same bug as the Phi-tiny-MoE autocast dtype mismatch." + "reason": "The titles suggest related loading trouble, but there is not enough evidence that they share the same tokenizer/vocab mismatch." }, { "left": "issue:43257", "right": "issue:44863", "accept": false, - "reason": "Different model families and failures: Qwen3 MoE weight conversion vs NemotronH checkpoint loading." + "reason": "Both are load failures, but for different model families and different loading pipelines." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Unrelated issues: missing Float8 storage type vs mmap-related OOM on Strix Halo." + "reason": "A Float8 storage TypeError and an mmap OOM workaround are unrelated failure modes." }, { - "left": "issue:44683", - "right": "issue:45823", + "left": "issue:39404", + "right": "issue:45701", "accept": false, - "reason": "Different subsystems: compiled flex_attention on torch>=2.9 vs Gemma4 PLE device-map mismatch." + "reason": "Whisper pipeline return_language regression is unrelated to a version-induced tokenization change in general." }, { - "left": "issue:39404", - "right": "issue:45701", + "left": "issue:44683", + "right": "issue:45823", "accept": false, - "reason": "Whisper pipeline return_language regression is unrelated to a general tokenization-version-change report." + "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 device-map forward mismatch are different problems." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "Torch 2.10 incompatibility is not the same concrete bug as Gemma4 PLE device mismatch." + "reason": "Torch compatibility and Gemma4 PLE device mismatch are not the same underlying bug." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "Both are Transformers 5 loading regressions, but one is base_model_prefix handling and the other is custom model initialization in notebooks." + "reason": "base_model_prefix loading failures and notebook-defined custom model initialization failures are different issues." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "No shared underlying bug; one is a framework compatibility complaint, the other is an image preprocessor/model dtype mismatch." + "reason": "Torch compatibility is unrelated to Mistral-3 image preprocessor/model dtype mismatch." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "PagedAttentionCache linear_attention crash is unrelated to the template-node compilation error." + "reason": "PagedAttentionCache linear_attention crash is unrelated to the non-template-node compile error." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "Isolated reproduction of a prior issue does not match the Qwen3.5 PagedAttentionCache crash." + "reason": "No clear shared bug beyond broad runtime-regression similarity." }, { "left": "issue:39404", "right": "issue:44912", "accept": false, - "reason": "Whisper return_language regression and MXFP4 quantization fallback are different behaviors and code paths." + "reason": "Whisper pipeline language return and MXFP4 quantization fallback are different issues." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "torch.export/torch_compilable_check export failure is unrelated to Perceiver interpolation failure." + "reason": "torch.export failures with torch_compilable_check are not the same as Perceiver interpolation failures." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Video input regression for Qwen VL models is not the same as the mask-shape IndexError." + "reason": "Video-input regression in Qwen VL models is unrelated to the mask/tensor shape IndexError." }, { - "left": "issue:43994", - "right": "issue:44079", + "left": "issue:43901", + "right": "issue:44509", "accept": false, - "reason": "Model output quality issue vs a ModelOutput key-assignment bug; different layers of the stack." + "reason": "Both are docs issues, but they refer to different deprecations and should not be merged." }, { - "left": "issue:43901", - "right": "issue:44509", + "left": "issue:43994", + "right": "issue:44079", "accept": false, - "reason": "Both are docs-related, but they refer to different pages and different deprecated/changed behaviors." + "reason": "A bad model output regression and ModelOutput key assignment behavior are distinct bugs." }, { "left": "issue:43425", "right": "issue:44918", "accept": false, - "reason": "Torch compatibility issue is unrelated to the TRL SFT trainer embedding unpacking problem." + "reason": "Torch 2.10 incompatibility is unrelated to Qwen3.5 input-embedding unpacking in TRL SFT." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Missing Float8 storage and template-node compilation are different failures." + "reason": "Different failures: missing Float8 storage object versus compile-time rejection of non-template nodes." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Qwen3VL MoE loading breakage and Qwen3-VL-30B-A3B shape mismatch are related only by model family, not the same bug." + "reason": "This is only supported by shared Qwen3-VL/model-family/title similarity; the packet does not show enough evidence of the same concrete bug or code path." }, { "left": "issue:44164", "right": "issue:45759", "accept": false, - "reason": "extra_state save/load failure is unrelated to from_config not unwrapping text_config." + "reason": "extra_state save/load handling is unrelated to unwrapping text_config for composite Qwen multimodal configs." }, { "left": "issue:36032", "right": "issue:43452", "accept": false, - "reason": "Tokenizer method conflict and gguf_file loading breakage are different bugs." - }, - { - "left": "issue:43596", - "right": "issue:44683", - "accept": false, - "reason": "Zero3/BertModel index error is unrelated to compiled flex_attention on newer torch." - }, - { - "left": "issue:44671", - "right": "issue:44991", - "accept": false, - "reason": "Different loading regressions for different models: CamemBERT MLM predictions vs EMBEDDIA tokenizer loading." - }, - { - "left": "issue:39404", - "right": "issue:44671", - "accept": false, - "reason": "Whisper pipeline behavior and CamemBERT masked-LM prediction errors are unrelated." - }, - { - "left": "issue:43856", - "right": "issue:44756", - "accept": false, - "reason": "Qwen3 MoE training memory usage and Strix Halo mmap OOM are not the same underlying issue." - }, - { - "left": "issue:44991", - "right": "issue:45081", - "accept": false, - "reason": "Tokenizer loading failure for EMBEDDIA is unrelated to the Mistral regex patch crash." - }, - { - "left": "issue:44164", - "right": "issue:44479", - "accept": false, - "reason": "extra_state persistence is unrelated to the Qwen VL video-input regression." - }, - { - "left": "issue:32090", - "right": "issue:39404", - "accept": false, - "reason": "Trainer gpu_broadcast_one NoneType error and Whisper return_language regression are unrelated." + "reason": "T5 tokenizer add_special_tokens conflict and gguf_file breaking from_pretrained are unrelated loading failures." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is grounded in the packet and stays conservative. The only accepted soft edge, 45292 \u2194 45335, is well supported by the closely matching titles about `resize_token_embeddings` not updating related embeddings. The rest of the pairwise rejections are cautious and do not overclaim shared root causes. No mergeability issue applies here because there are no PR pairs." + "accept": false, + "feedback": "The summary is mostly grounded, but one soft-edge verdict is too strong for the packet: issue:43299 vs issue:43931 is only supported by shared model family/title similarity, not clear evidence of the same underlying bug. Calling it the same loading problem overstates what the packet shows. The resize_token_embeddings pair (45292/45335) looks conservative and well-supported, but the Qwen3-VL pair should be treated as merely plausible, not accepted as a duplicate." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -2650,22 +2522,21 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ + "issue:32090", "issue:34567", "issue:36032", + "issue:39404", "issue:40444", "issue:40990", "issue:42175", "issue:42915", - "issue:43089", - "issue:43262", "issue:43317", - "issue:43377", - "issue:43404", "issue:43421", "issue:43452", "issue:43482", "issue:43541", "issue:43582", + "issue:43596", "issue:43611", "issue:43632", "issue:43643", @@ -2673,18 +2544,21 @@ "issue:43645", "issue:43646", "issue:43716", + "issue:43856", "issue:43883", + "issue:44164", "issue:44206", "issue:44220", "issue:44451", "issue:44464", - "issue:44466", + "issue:44479", "issue:44488", "issue:44530", - "issue:44534", "issue:44661", + "issue:44671", "issue:44683", "issue:44737", + "issue:44756", "issue:44811", "issue:44849", "issue:44987", @@ -2692,13 +2566,19 @@ "issue:45081", "issue:45084", "issue:45325", - "issue:45507", "issue:45584", "issue:45859", "issue:45987", "issue:46178" ], "soft_pairs": [ + "issue:43596|issue:44683", + "issue:44671|issue:44991", + "issue:39404|issue:44671", + "issue:44991|issue:45081", + "issue:43856|issue:44756", + "issue:44164|issue:44479", + "issue:32090|issue:39404", "issue:44737|issue:45859", "issue:43582|issue:43632", "issue:44661|issue:44683", @@ -2711,8 +2591,8 @@ "issue:36032|issue:42915", "issue:43883|issue:44488", "issue:43643|issue:44530", - "issue:42175|issue:46178", "issue:34567|issue:40444", + "issue:42175|issue:46178", "issue:44811|issue:45584", "issue:43541|issue:45084", "issue:44451|issue:44987", @@ -2723,242 +2603,238 @@ "issue:36032|issue:43645", "issue:43317|issue:43482", "issue:44849|issue:45325", - "issue:44987|issue:44991", - "issue:45084|issue:45507", - "issue:43883|issue:44991", - "issue:43262|issue:43377", - "issue:43089|issue:45987", - "issue:43645|issue:44534", - "issue:43646|issue:44534", - "issue:43404|issue:44466" + "issue:44987|issue:44991" ], - "prepared_review_unit_hash": "b53d5cf94fded375bfe0f1670f2918189b7c2fc2241441400baf53ffb57fa051", + "prepared_review_unit_hash": "0e3e6332b3dbb2637b15432ab57bc959f1c045393a6a94dfefd340eeff2cc456", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36827, - "estimated_input_tokens": 9207, - "estimated_eval_tokens": 18670 + "serialized_chars": 37234, + "estimated_input_tokens": 9309, + "estimated_eval_tokens": 18874 }, "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 and model-loading bugs. The titles share broad topical overlap, but there is no clear single duplicate center, and none of the soft edges look like the same concrete bug or change.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue stands out: the items span several different regressions (tokenizers, model loading, audio, generation, caching, and v5 initialization), so this does not look like a true duplicate cluster.", + "summary": "This cluster is mostly a set of loosely related Transformers v5 regression reports. I only see one clear duplicate pair: the audio feature-extraction/fbank crash. The rest are thematically similar at best, but not the same concrete bug.", + "confidence": 0.67, + "canonical_issue_reason": "issue:44206 is the better umbrella report for the fbank/feature-extraction crash; issue:44220 looks like the same code path expressed through the helper-level failure.", "canonical_pr_reason": null, - "best_issue_reason": "If one issue had to serve as a representative, #43644 (non-persistent buffers filled with junk in v5) is the closest broad Transformers v5 regression anchor, but it is still only a theme representative, not a dedupe canonical.", + "best_issue_reason": "issue:44206 is the most representative and concrete issue in the only clear duplicate pair, with the broadest description of the underlying audio regression.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Different failures: deepspeed/zero3 index error in BertModel vs a torch 2.9 flex_attention compile regression." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading-related, but one is a CamemBERT prediction regression and the other is a tokenizer loading failure for a different model." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline return_language regression vs CamemBERT masked-LM output bug; unrelated code paths." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both touch tokenizer loading, but the models and failure modes differ; no evidence of the same underlying bug." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3 MoE training memory usage vs Strix Halo mmap OOM; same general theme, different causes." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "Model save/load extra_state handling vs a video-input regression; not the same bug." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error and Whisper return_language pipeline failure are unrelated." + }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both are device-placement bugs, but they hit different models and code paths (XLNet arange device vs Qwen3.5MoE rotary embedding CPU offload)." + "reason": "Both involve device placement, but they affect different models and different forward paths." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "Unrelated regressions: Apple Silicon allocator warmup TypeError vs the _is_hf_initialized flag breakage." + "reason": "Apple Silicon allocator warmup TypeError vs _is_hf_initialized flag regression; no shared concrete bug." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Different subsystems entirely: tokenizer mapping validation vs compiled flex_attention failing on newer torch." + "reason": "Tokenizer mapping/model-addition logic vs compiled flex_attention tracing; unrelated." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer-loading failures, but the specific failure modes and affected paths differ too much to be the same bug." + "reason": "Both are tokenizer-loading complaints, but about different models and different breakages." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "These are separate model-loading failures on different models with no clear shared code-path bug." + "reason": "Two different model-loading failures; only a broad transformer-version theme is shared." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Base-model-prefix loading regression and non-persistent buffer junk are distinct v5 issues, not one bug." + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity regression on a model/dataset is unrelated to runtime special-token post-processor updates." + "reason": "Perplexity regression and tokenizer post-processor update bug are different issues." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Image-preprocessor dtype mismatch and template compilation errors are unrelated." + "reason": "Image preprocessor dtype mismatch vs template compilation TypeError; unrelated." }, { "left": "issue:44206", "right": "issue:44220", - "accept": false, - "reason": "Both touch audio features, but one is an unsupported argument in a feature extractor and the other is an internal fbank helper issue." + "accept": true, + "reason": "Same audio feature-extraction path: one reports the unsupported center arg crash, the other reports the downstream _torch_extract_fbank_features failure." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Tokenizer method-name conflict and Qwen3Moe FP8 failure are unrelated." + "reason": "T5 tokenizer add_special_tokens conflict vs Qwen3Moe FineGrainedFP8Config failure; unrelated." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Missing all_tied_weights_keys and a model-loading failure for a specific BERT variant are different issues." + "reason": "Missing all_tied_weights_keys on Molmo vs a model-loading regression for a different repository." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "trust_remote_code config field loss is unrelated to PagedAttentionCache's invalid group type error." + "reason": "AutoConfig trust_remote_code field loss vs Qwen3.5 cache group-type crash; different code paths." }, { - "left": "issue:42175", - "right": "issue:46178", + "left": "issue:34567", + "right": "issue:40444", "accept": false, - "reason": "Both concern backend handling, but one is missing tensorflow from an extra and the other is an unsupported backend mapping entry." + "reason": "TrainerState token counter not updating vs Qwen2.5-VL iterable-dataset multi-image finetuning failure." }, { - "left": "issue:34567", - "right": "issue:40444", + "left": "issue:42175", + "right": "issue:46178", "accept": false, - "reason": "Trainer token counting and multimodal iterable-dataset failures are unrelated." + "reason": "Both are backend-installation complaints, but tensorflow missing from torch extras and tensorflow_text backend mapping are distinct issues." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both involve Whisper special tokens, but the bugs are different: batch_decode ignoring skip_special_tokens vs empty transcription after align_special_tokens." + "reason": "Both involve Whisper special-token behavior, but one is batch_decode skipping tokens and the other is empty-transcription generation after alignment." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Torch dynamo grouped_mm tracing failure and a non-template-node compile error are unrelated." + "reason": "Torch dynamo grouped_mm tracing error vs template-node compilation error; not the same bug." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Different model-loading regressions on different models; same broad symptom, not the same bug." + "reason": "Different model-loading failures; no concrete shared code-path shown." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "Both are Qwen-family issues, but FP8 config failure and linear_attention cache-group failure are distinct code paths." + "reason": "Qwen3Moe FP8 config failure vs Qwen3.5 PagedAttentionCache group-type crash; related model family, different bugs." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Related v5 initialization theme, but one is buffer-state corruption and the other is custom-model initialization failure; not the same concrete bug." + "reason": "Both are v5 regressions, but one concerns buffer filling and the other custom model initialization." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Chunked generation inconsistency under compiled forward and a seq_length return-type contract bug are not the same issue." + "reason": "Compiled chunked generation inconsistency vs StaticCache return-type contract bug; different failures." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "Both mention GGUF, but one is a broad from_pretrained gguf_file regression and the other is a specific Qwen2.5-GGUF load failure." + "reason": "Likely both concern GGUF loading, but the titles do not show the same concrete bug/change with enough confidence." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "Tokenizer special-token method conflict and custom model init in notebooks are unrelated." + "reason": "Tokenizer special-token conflict vs Jupyter custom-model initialization breakage; unrelated." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "Dequantized model offload loading and GGUF loading are different mechanisms and failure modes." + "reason": "GPU+CPU offload dequantized-model load failure vs Qwen2.5-GGUF loading failure; not enough evidence of one bug." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Qwen3.5 hidden-states behavior and Qwen2.5-VL rope index scaling are separate bugs." + "reason": "Qwen3.5 hidden-states bug vs Qwen2.5-VL rope-index scaling bug; different symptoms and code paths." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both are versioned loading failures, but on different repositories/models with no evidence of a shared root cause." - }, - { - "left": "issue:45084", - "right": "issue:45507", - "accept": false, - "reason": "Template compilation and GraniteMoEHybrid calling an invalid method are unrelated." - }, - { - "left": "issue:43883", - "right": "issue:44991", - "accept": false, - "reason": "Missing tied-weights metadata and tokenizer-loading failure are unrelated." - }, - { - "left": "issue:43262", - "right": "issue:43377", - "accept": false, - "reason": "Audio sampling-rate defaulting and MIMI encoder padding-mask behavior affect different layers of the stack." - }, - { - "left": "issue:43089", - "right": "issue:45987", - "accept": false, - "reason": "Generation performance overhead and cache sequence-length return type are different problems." - }, - { - "left": "issue:43645", - "right": "issue:44534", - "accept": false, - "reason": "Custom model initialization in notebooks is not the same bug as junk non-persistent buffers in v5." - }, - { - "left": "issue:43646", - "right": "issue:44534", - "accept": false, - "reason": "Custom model initialization breakage and non-persistent buffer corruption are separate regressions." - }, - { - "left": "issue:43404", - "right": "issue:44466", - "accept": false, - "reason": "Tied-weight handling in Mistral3ForConditionalGeneration and device-dependent lm_head serialization are related in theme only, not the same concrete defect." + "reason": "Both are loading regressions on different models, but the titles do not establish a shared underlying fixable bug." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays mostly conservative. The one accepted soft pair, issue:44206 vs issue:44220, is a reasonable duplicate call based on the shared audio/fbank feature-extraction path, and the other soft-edge rejections are cautious and supported by the titles." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -2975,56 +2851,62 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ - "issue:33666", "issue:36032", "issue:40444", - "issue:40990", "issue:41553", "issue:41720", "issue:42915", + "issue:43089", + "issue:43262", "issue:43296", "issue:43322", "issue:43377", "issue:43404", - "issue:43421", "issue:43425", "issue:43531", "issue:43540", "issue:43582", "issue:43611", "issue:43632", + "issue:43645", + "issue:43646", "issue:43742", - "issue:43792", "issue:43819", + "issue:43883", "issue:43994", "issue:44062", - "issue:44079", "issue:44220", "issue:44265", "issue:44291", "issue:44403", "issue:44451", + "issue:44466", "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:45663" + "issue:45507", + "issue:45987" ], "soft_pairs": [ + "issue:45084|issue:45507", + "issue:43883|issue:44991", + "issue:43262|issue:43377", + "issue:43645|issue:44534", + "issue:43089|issue:45987", + "issue:43646|issue:44534", + "issue:43404|issue:44466", "issue:43322|issue:43540", "issue:44403|issue:45081", "issue:43994|issue:44610", @@ -3037,247 +2919,240 @@ "issue:42915|issue:44291", "issue:45216|issue:45357", "issue:43632|issue:44756", - "issue:44062|issue:44987", "issue:36032|issue:43742", + "issue:44062|issue:44987", "issue:44479|issue:45325", "issue:42915|issue:44589", - "issue:43404|issue:45127", "issue:43296|issue:43531", + "issue:43404|issue:45127", "issue:43531|issue:44877", - "issue:40444|issue:45325", "issue:36032|issue:41720", + "issue:40444|issue:45325", "issue:43611|issue:44534", "issue:43425|issue:44403", - "issue:41553|issue:43582", "issue:43632|issue:44534", - "issue:40990|issue:44811", - "issue:44079|issue:45663", - "issue:43792|issue:44589", - "issue:44683|issue:45020", - "issue:33666|issue:40444", - "issue:44403|issue:44991", - "issue:43421|issue:44568" + "issue:41553|issue:43582" ], - "prepared_review_unit_hash": "d7eb5f808d7786cf7d3622cd71ae61019d5a524dfc8c6a77311e9f8ed1c83fcb", + "prepared_review_unit_hash": "20442020b040489ed9cf611af12c6355644f6b57212beab2de63cbbb7d3627e2", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 37178, - "estimated_input_tokens": 9295, - "estimated_eval_tokens": 18846 + "serialized_chars": 36679, + "estimated_input_tokens": 9170, + "estimated_eval_tokens": 18596 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is loosely centered on Transformers loading/regression reports, but the individual issues are mostly distinct bugs across different models and code paths. None of the soft pairs look like true duplicates, so they should stay separate.", - "confidence": 0.58, - "canonical_issue_reason": "issue:45020 is the broadest representative of the cluster: a recent-version regression breaking model loading with remote code, which matches the dominant loading/regression theme better than the narrower model-specific reports.", + "summary": "The cluster is broadly about Transformers v5-era regressions, but the individual issues span many unrelated subsystems and model families. I found no PRs, and none of the soft edges look like safe duplicates under a strict standard.", + "confidence": 0.44, + "canonical_issue_reason": "issue:43611 is the broadest, most representative loading-regression report in the set; it best fits the cluster\u2019s loose \"Transformers v5 breaks model loading\" theme, even though the cluster is heterogeneous.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the best issue to represent the cluster because it captures the widest upstream regression pattern and can serve as the umbrella reference for related loading breakages.", + "best_issue_reason": "issue:43611 is the most suitable umbrella issue among these because it is concrete, widely scoped, and closest to the recurring v5 loading failures represented here.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Different problems: template-node compilation vs an invalid method call in GraniteMoEHybrid." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Unrelated failures: missing all_tied_weights_keys on Molmo vs tokenizer loading for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Different domains and bugs: audio chat-template sampling rate vs MIMI encoder padding-mask mismatch." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Custom model init in notebooks is not the same as non-persistent buffers being filled with junk." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation performance overhead and StaticCache return-type contract are distinct issues." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Both are v5 regressions, but they concern different code paths: custom model init vs buffer serialization/init behavior." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both mention tied weights, but one is missing weight tying in model init while the other is save_pretrained serialization behavior." + }, { "left": "issue:43322", "right": "issue:43540", "accept": false, - "reason": "Different failures: Llava Next loading segfault vs Qwen3OmniMoe video-input ValueError." + "reason": "Different model families and failure modes: Llava Next segfault vs Qwen3OmniMoe video-processing ValueError." }, { "left": "issue:44403", "right": "issue:45081", "accept": false, - "reason": "Generic loading-noise complaint vs a specific Mistral regex crash in tokenizer patching." + "reason": "Generic loading noise is unrelated to a Mistral regex patch crash." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Different model families and symptoms: SigLIP2 bad outputs vs OmDet-Turbo processor/model size mismatch." + "reason": "Different models and symptoms: Siglip2 bad outputs vs OmDet-Turbo processor size mismatch." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "Unrelated bugs: Qwen3Moe FP8 config failure vs template compilation TypeError." + "reason": "FineGrainedFP8 model failure is unrelated to a template-node compilation TypeError." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "Both are audio-related, but one is MIMI padding-mask inconsistency and the other is DAC latent/STE mismatch." + "reason": "Both are encoder/model bugs, but the concrete failures differ: padding-mask batching vs DAC latent reconstruction." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export/torch_compilable_check failure is a different code path from the OmDet-Turbo input-size regression." + "reason": "torch.export failure with torch_compilable_check is unrelated to OmDet processor input resolution." }, { "left": "issue:44451", "right": "issue:45081", "accept": false, - "reason": "Different models and root causes: ScandiBERT loading failure vs Mistral regex patch crash." + "reason": "Model load failure for ScandiBERT and Mistral tokenizer regex crash are different issues." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "fast model load regression is unrelated to the Mistral tokenizer backend_tokenizer error." + "reason": "Loading physical-intelligence/fast on v5.1.0 is a different regression from the Mistral tokenizer patch crash." }, { "left": "issue:44220", "right": "issue:44683", "accept": false, - "reason": "Audio feature extraction bug vs compiled flex_attention failure on newer torch." + "reason": "Audio feature extraction and compiled flex_attention on torch>=2.9 are separate code paths." }, { "left": "issue:42915", - "right": "issue:44589", + "right": "issue:44291", "accept": false, - "reason": "Qwen3Moe FP8 config issue is unrelated to the Float8 storage lookup/type error." + "reason": "FP8 config failure and init_empty_weights/_is_hf_initialized argument handling are different bugs." }, { "left": "issue:45216", "right": "issue:45357", "accept": false, - "reason": "Related to Qwen3.5 save_pretrained regressions, but they target different save bugs: checkpoint correctness vs visual encoder keys." + "reason": "Related theme, but not clearly the same bug: general Qwen3.5 checkpoint corruption vs visual-encoder key serialization specifically." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "Transformers v5 initialization-flag regression vs a Strix Halo mmap OOM workaround." + "reason": "_is_hf_initialized breakage and mmap OOM on Strix Halo are unrelated." }, { - "left": "issue:44062", - "right": "issue:44987", + "left": "issue:36032", + "right": "issue:43742", "accept": false, - "reason": "Tokenizer AddedToken argument conflict is not the same as loading physical-intelligence/fast failing." + "reason": "Tokenizer method conflict in T5 and a key error loading MobileLLM are not the same underlying issue." }, { - "left": "issue:36032", - "right": "issue:43742", + "left": "issue:44062", + "right": "issue:44987", "accept": false, - "reason": "Tokenizer constructor conflict vs a key error when loading a specific model." + "reason": "AddedToken special-argument conflict and a model-loading regression for physical-intelligence/fast are distinct." }, { "left": "issue:44479", "right": "issue:45325", "accept": false, - "reason": "Both concern Qwen multimodal models, but one is a video-input regression and the other is a still-image rope-index scaling bug." + "reason": "Both involve Qwen vision models, but one is a video-input regression and the other is still-image temporal position scaling." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "Different failure modes and subsystems: FP8 config handling vs Float8 storage lookup." + "reason": "Both touch FP8-ish areas, but one is a config/load failure and the other is a missing Float8 storage type." }, { - "left": "issue:43404", - "right": "issue:45127", + "left": "issue:43296", + "right": "issue:43531", "accept": false, - "reason": "Weight-tying bug in Mistral3 conditional generation is unrelated to LoRA merge collapse with extended vocab." + "reason": "PaddleOCR-VL loading and Qwen3-MoE sliding_window behavior are unrelated." }, { - "left": "issue:43296", - "right": "issue:43531", + "left": "issue:43404", + "right": "issue:45127", "accept": false, - "reason": "PaddleOCR-VL loading failure and Qwen3-MoE sliding_window behavior are different model-specific bugs." + "reason": "Both involve tied weights, but one is a model class not tying lm_head while the other is LoRA merge collapse with extended vocabulary." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Qwen3-MoE sliding_window issue is unrelated to granite_speech strict-config loading failure." + "reason": "Qwen3-MoE sliding_window bug and granite_speech strict-config loading failure are unrelated." }, { - "left": "issue:40444", - "right": "issue:45325", + "left": "issue:36032", + "right": "issue:41720", "accept": false, - "reason": "Different Qwen2.5-VL bugs: iterable dataset multi-image finetuning vs rope-index temporal scaling." + "reason": "Tokenizer loading conflict in T5 and Qwen3 auto device-mapping CUDA assert are different failures." }, { - "left": "issue:36032", - "right": "issue:41720", + "left": "issue:40444", + "right": "issue:45325", "accept": false, - "reason": "Tokenizer load conflict is unrelated to Qwen3 auto device mapping cudaErrorAssert." + "reason": "IterableDataset multi-image finetuning and Qwen2.5-VL rope-index scaling are different bugs." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "Base_model_prefix loading breakage is a different issue than non-persistent buffers being filled with junk." + "reason": "Base_model_prefix loading regression and non-persistent buffer junk are separate v5 issues." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch version incompatibility is not the same as generic transformer-load noise." - }, - { - "left": "issue:41553", - "right": "issue:43582", - "accept": false, - "reason": "AutoTokenizer error for Voxtral and Apple Silicon warmup TypeError are unrelated." + "reason": "Torch version incompatibility and generic loading noise are not the same bug." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "_is_hf_initialized regression is distinct from the non-persistent buffer corruption bug." - }, - { - "left": "issue:40990", - "right": "issue:44811", - "accept": false, - "reason": "High perplexity on GPT-OSS and Whisper batch_decode skip_special_tokens handling are different problems." - }, - { - "left": "issue:44079", - "right": "issue:45663", - "accept": false, - "reason": "ModelOutput key assignment bug is unrelated to Gemma-4 FSDP2 shared_kv_states rebuild failure." - }, - { - "left": "issue:43792", - "right": "issue:44589", - "accept": false, - "reason": "Whisper model runtime failure is not the same as the Float8 storage-object TypeError." - }, - { - "left": "issue:44683", - "right": "issue:45020", - "accept": false, - "reason": "Compiled flex_attention torch-version failure is unrelated to the remote_code loading regression." - }, - { - "left": "issue:33666", - "right": "issue:40444", - "accept": false, - "reason": "Both are Qwen2-VL/Qwen2.5-VL, but one is multi-GPU training and the other is iterable-dataset multi-image finetuning." - }, - { - "left": "issue:44403", - "right": "issue:44991", - "accept": false, - "reason": "Generic loading-noise report and est-roberta tokenizer loading failure are not the same bug." + "reason": "_is_hf_initialized handling and non-persistent buffer initialization are different code paths." }, { - "left": "issue:43421", - "right": "issue:44568", + "left": "issue:41553", + "right": "issue:43582", "accept": false, - "reason": "Runtime post-processor update behavior is different from missing BOS/EOS insertion in mdeberta-v3-base." + "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to the Apple Silicon caching_allocator_warmup TypeError." } ] }, @@ -3301,16 +3176,17 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ + "issue:33666", "issue:35707", "issue:40444", + "issue:40990", "issue:41720", "issue:42175", - "issue:42915", "issue:43296", "issue:43317", "issue:43377", "issue:43381", - "issue:43454", + "issue:43421", "issue:43482", "issue:43525", "issue:43531", @@ -3318,24 +3194,24 @@ "issue:43632", "issue:43644", "issue:43646", - "issue:43701", "issue:43749", "issue:43792", "issue:43818", "issue:43856", "issue:43957", - "issue:43994", "issue:44079", "issue:44265", "issue:44291", - "issue:44387", + "issue:44403", "issue:44462", - "issue:44466", "issue:44534", + "issue:44568", "issue:44589", - "issue:44617", + "issue:44683", + "issue:44811", "issue:44829", - "issue:44898", + "issue:44991", + "issue:45020", "issue:45072", "issue:45081", "issue:45325", @@ -3345,25 +3221,31 @@ "issue:45491", "issue:45538", "issue:45542", - "issue:45800", + "issue:45663", "issue:45812", - "issue:45834", "issue:45923" ], "soft_pairs": [ + "issue:40990|issue:44811", + "issue:44079|issue:45663", + "issue:43792|issue:44589", + "issue:33666|issue:40444", + "issue:44683|issue:45020", + "issue:44403|issue:44991", + "issue:43421|issue:44568", "issue:43381|issue:45923", "issue:35707|issue:40444", "issue:44462|issue:45812", - "issue:42175|issue:45542", - "issue:43317|issue:43856", - "issue:43482|issue:45538", "issue:44291|issue:44589", + "issue:43482|issue:45538", + "issue:42175|issue:45542", "issue:44829|issue:45479", + "issue:43317|issue:43856", "issue:43482|issue:43525", "issue:44079|issue:44534", "issue:43818|issue:45357", - "issue:43377|issue:45491", "issue:43531|issue:44291", + "issue:43377|issue:45491", "issue:41720|issue:45081", "issue:40444|issue:45381", "issue:44265|issue:45072", @@ -3373,229 +3255,222 @@ "issue:45325|issue:45381", "issue:43632|issue:45072", "issue:43611|issue:43646", - "issue:43792|issue:43957", - "issue:45325|issue:45834", - "issue:43994|issue:44898", - "issue:43454|issue:44466", - "issue:42915|issue:43482", - "issue:43296|issue:45800", - "issue:44387|issue:44617", - "issue:43701|issue:45325" + "issue:43792|issue:43957" ], - "prepared_review_unit_hash": "a45f71bfda3b1ee07c5750a677fd1086fb8b03e46c157edb2d6c805dddeeee61", + "prepared_review_unit_hash": "60f98f8f3e0f69d4a3c5c4b0ea97015f5970f78272763cf723dbd97719939ac1", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36846, - "estimated_input_tokens": 9212, - "estimated_eval_tokens": 18680 + "serialized_chars": 36902, + "estimated_input_tokens": 9226, + "estimated_eval_tokens": 18708 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly noisy Transformers-v5 regression reports, with only a few true duplicate pairs. The strongest shared themes are initialization/loading regressions, TF backend packaging, and Qwen2.5-VL position-id bugs.", - "confidence": 0.64, - "canonical_issue_reason": "Issue 43632 is the most central representative of the cluster\u2019s main theme: a broad Transformers v5 initialization/loading regression that can plausibly underlie several related model-loading complaints.", + "summary": "This cluster is mostly heterogeneous; most soft pairs are related by broad subsystem/topic only, not the same bug. Two pairs do look like genuine duplicates: the AutoTokenizer/tokenizer.json regression and the Qwen2.5-VL position-id regression. I\u2019d use the tokenizer-loading issue as the main representative.", + "confidence": 0.61, + "canonical_issue_reason": "issue:44462 is the broadest concrete bug here: AutoTokenizer ignoring tokenizer.json. It also cleanly subsumes the Granite-specific symptom in issue:45812, making it the best umbrella issue in this set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43632 is the best overall anchor because it captures the broadest and most reusable failure mode in the cluster, rather than a model-specific symptom.", + "best_issue_reason": "issue:44462 is the best representative because it is a clear, root-cause tokenizer-loading regression rather than a model-specific symptom, and it matches the strongest duplicate pair in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Different problems and code paths: perplexity evaluation vs Whisper batch_decode special-token handling." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment bug vs Gemma-4 FSDP shared_kv_states rebuild; unrelated." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure and Float8 storage lookup error are different failures." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are Qwen VL training issues, but the failures are different: multi-GPU training vs iterable dataset with multiple images." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and remote_code model loading regressions are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic loading noise vs tokenizer loading failure; not the same underlying bug." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Both involve tokenizer special tokens, but one is runtime post-processor refresh and the other is BOS/EOS insertion for a specific tokenizer." + }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Different problems: eval-mode checkpointing behavior vs a missing supports_gradient_checkpointing flag on a remote-code model." + "reason": "Eval-mode gradient checkpointing misuse is not the same as a missing supports_gradient_checkpointing flag." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Unrelated symptoms and code paths: generation with inputs_embeds/past_key_values vs multimodal finetuning with multiple images per prompt." + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen VL finetuning." }, { "left": "issue:44462", "right": "issue:45812", - "accept": false, - "reason": "Both involve tokenizers, but one is repository file selection and the other is wrong token IDs for Granite models; different bugs." - }, - { - "left": "issue:42175", - "right": "issue:45542", "accept": true, - "reason": "Both point to the same TensorFlow/extra-dependency packaging issue causing TF backend availability errors." + "reason": "Both point to the same AutoTokenizer loading regression; Granite wrong token IDs plausibly come from ignoring tokenizer.json." }, { - "left": "issue:43317", - "right": "issue:43856", + "left": "issue:44291", + "right": "issue:44589", "accept": false, - "reason": "Different failures: dequantized model loading with device_map/offload vs MoE training memory usage." + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage object are different init/load failures." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "Different tokenizer/model issues: GGUF loading regression vs CLIPTokenizer model_max_length misconfiguration." + "reason": "Qwen2.5-GGUF loading failure and CLIPTokenizer model_max_length are unrelated tokenizer/model-loading bugs." }, { - "left": "issue:44291", - "right": "issue:44589", + "left": "issue:42175", + "right": "issue:45542", "accept": false, - "reason": "Distinct v5 regressions: init_empty_weights/_is_hf_initialized argument handling vs missing Float8 storage lookup." + "reason": "Both touch TensorFlow packaging, but one is missing backend in the torch extra and the other is a tf-backend detection/runtime issue." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Both are training pathologies, but one is flash_attention_3 causing degenerate training and the other is a label/loss edge case with num_labels=1." + "reason": "Flash-attention-3 training degeneration and single-label num_labels=1 zero-loss are different classification bugs." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Device-map offload loading failure and Qwen3 MoE memory inefficiency are distinct issues." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "No shared underlying bug: GGUF loading failure vs Llama4Config missing pad_token_id." + "reason": "Different model/config failures: GGUF loading vs missing pad_token_id on Llama4Config." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "Different regressions: ModelOutput key assignment vs non-persistent buffer serialization." + "reason": "ModelOutput key assignment and non-persistent buffer junk are separate core regressions." }, { "left": "issue:43818", "right": "issue:45357", "accept": false, - "reason": "Different multimodal issues affecting different models and code paths: Video-LLaVA tower weights vs Qwen3.5 visual-encoder save_pretrained keys." + "reason": "Video-LLaVA tower weights/temporal attention issue is unrelated to Qwen3.5 save_pretrained key regression." }, { - "left": "issue:43377", - "right": "issue:45491", + "left": "issue:43531", + "right": "issue:44291", "accept": false, - "reason": "Both involve batching, but one is missing padding-mask support in MIMI and the other is Gemma3 NaNs from sliding-window attention." + "reason": "Sliding-window behavior in Qwen3-MoE and init_empty_weights argument handling are unrelated." }, { - "left": "issue:43531", - "right": "issue:44291", + "left": "issue:43377", + "right": "issue:45491", "accept": false, - "reason": "Unrelated: Qwen3-MoE sliding_window behavior vs init_empty_weights argument handling." + "reason": "Both concern batching/padding, but they are different model-specific bugs with different symptoms." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Different subsystems and failures: Qwen3 auto device mapping CUDA assert vs Mistral tokenizer regex patch crash." + "reason": "Qwen3 auto device-map assert and Mistral tokenizer regex patch crash are unrelated." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is finetuning with multiple images and the other is video position-id computation; not the same bug." + "reason": "Both are Qwen2.5-VL, but one is a finetuning/dataset failure and the other is a video position-id bug." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "No common code path: torch.export with torch_compilable_check vs bfloat16 dtype mismatches in inference." + "reason": "torch.export/torch_compilable_check failure and bfloat16 dtype mismatch are unrelated." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Different v5 regressions: non-persistent buffer junk vs ModelOutput key bookkeeping." + "reason": "Non-persistent buffer junk and ModelOutput key assignment are different bugs." }, { "left": "issue:43749", "right": "issue:44079", "accept": false, - "reason": "Different areas: FSDP CPU RAM efficient loading vs ModelOutput key assignment." + "reason": "FSDP CPU RAM efficient loading regression is unrelated to ModelOutput key assignment." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Different topics: TensorFlow backend installation vs PaddleOCR-VL model loading in vLLM." + "reason": "TensorFlow extra/backend packaging and PaddleOCR-VL model loading are different issues." }, { "left": "issue:45325", "right": "issue:45381", "accept": true, - "reason": "Same underlying Qwen2.5-VL position-id/rope-index bug across still-image and video inputs." + "reason": "Both describe the same Qwen2.5-VL position-id calculation regression affecting image/video paths in the same area of code." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "Different failure modes: model initialization flag regression vs bfloat16 dtype mismatch in inference." + "reason": "_is_hf_initialized regression and bfloat16 dtype mismatch are unrelated." }, { "left": "issue:43611", "right": "issue:43646", - "accept": true, - "reason": "Both are Transformers v5 model-initialization/load regressions affecting custom-model handling and could plausibly share the same root fix." + "accept": false, + "reason": "Both are transformers v5 regressions, but base_model_prefix loading and custom model initialization are not clearly the same bug." }, { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Different loading/runtime problems: Whisper cannot run vs meta-device loading breakage on some models." - }, - { - "left": "issue:45325", - "right": "issue:45834", - "accept": false, - "reason": "Same broad family only by vague OCR/long-input failure; the models and concrete bug reports are different." - }, - { - "left": "issue:43994", - "right": "issue:44898", - "accept": false, - "reason": "Different vision-model issues with different causes: Siglip2 bad outputs vs Perceiver resolution/interpolation failure." - }, - { - "left": "issue:43454", - "right": "issue:44466", - "accept": false, - "reason": "Both mention lm_head/tied weights, but one is broken weight tying causing bad generation and the other is device-dependent serialization inconsistency." - }, - { - "left": "issue:42915", - "right": "issue:43482", - "accept": false, - "reason": "Different model families and failures: Qwen3Moe FP8 config vs Qwen2.5-GGUF loading." - }, - { - "left": "issue:43296", - "right": "issue:45800", - "accept": false, - "reason": "Unrelated: PaddleOCR-VL loading in vLLM vs torch/transformers version incompatibility." - }, - { - "left": "issue:44387", - "right": "issue:44617", - "accept": false, - "reason": "Different performance bugs: int4 CUDA reserved-memory growth vs Sam3Video OOM." - }, - { - "left": "issue:43701", - "right": "issue:45325", - "accept": false, - "reason": "No shared underlying bug: resume_from_checkpoint key mismatch vs Qwen2.5-VL position-id scaling." + "reason": "Whisper can\u2019t run and torch.device('meta') loading regressions are different failures." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded overall. The cluster is correctly characterized as mostly noisy Transformers v5 regression issues, and the selected anchor/canonical issue (43632) is a reasonable conservative representative of the broader initialization/loading regression theme. The accepted soft pairs are also cautious and supported by the issue titles: 42175\u219445542 on TF backend packaging, 45325\u219445381 on Qwen2.5-VL position-id/rope handling, and 43611\u219443646 on model initialization/loading regressions. I don\u2019t see any soft-edge verdict that overstates the evidence from the packet." + "feedback": "The summary is grounded in the packet overall, and the two accepted soft pairs are reasonably conservative based on the issue titles and subsystem overlap. The representative/canonical choice of issue:44462 is also defensible for this cluster." }, "evaluator_used": true, "retried": false, @@ -3622,6 +3497,7 @@ "issue:40444", "issue:40990", "issue:41720", + "issue:42915", "issue:42947", "issue:43257", "issue:43262", @@ -3630,49 +3506,55 @@ "issue:43322", "issue:43377", "issue:43454", + "issue:43482", "issue:43526", "issue:43550", "issue:43572", "issue:43606", - "issue:43720", + "issue:43701", "issue:43746", - "issue:43749", "issue:43819", "issue:43873", "issue:43881", "issue:43883", + "issue:43994", "issue:44186", "issue:44265", "issue:44291", "issue:44466", "issue:44479", "issue:44488", - "issue:44534", "issue:44568", "issue:44610", "issue:44877", "issue:44898", - "issue:44977", - "issue:44991", "issue:45072", - "issue:45357", + "issue:45325", "issue:45584", "issue:45701", "issue:45759", + "issue:45800", "issue:45823", + "issue:45834", "issue:45920", "issue:45923", "issue:46050" ], "soft_pairs": [ + "issue:45325|issue:45834", + "issue:43994|issue:44898", + "issue:43454|issue:44466", + "issue:42915|issue:43482", + "issue:43296|issue:45800", + "issue:43701|issue:45325", "issue:44479|issue:45823", "issue:44291|issue:44877", "issue:43881|issue:44488", "issue:43606|issue:44265", "issue:43873|issue:45701", "issue:43262|issue:43746", - "issue:34634|issue:35707", "issue:40990|issue:45584", + "issue:34634|issue:35707", "issue:44568|issue:45920", "issue:35707|issue:41720", "issue:44186|issue:44265", @@ -3683,234 +3565,203 @@ "issue:43377|issue:44610", "issue:43454|issue:45072", "issue:43572|issue:45759", - "issue:43317|issue:43873", "issue:43257|issue:45923", + "issue:43317|issue:43873", "issue:43377|issue:44898", - "issue:43526|issue:43883", "issue:40990|issue:42947", - "issue:43322|issue:46050", - "issue:43881|issue:44991", - "issue:43454|issue:43720", - "issue:41720|issue:43322", - "issue:43749|issue:44534", - "issue:43296|issue:44877", - "issue:44466|issue:45357", - "issue:41720|issue:44977" + "issue:43526|issue:43883", + "issue:43322|issue:46050" ], - "prepared_review_unit_hash": "ff1ac8116a57eab8aa8e6ff3cd12bd00dfa51c103fa1413a33a175248dcc4ac9", + "prepared_review_unit_hash": "8fda98609b42870fc40d8e0a1655411c72060335a3e06f3031b451b7ffc18eed", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 31, - "serialized_chars": 36633, - "estimated_input_tokens": 9159, - "estimated_eval_tokens": 18574 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36832, + "estimated_input_tokens": 9208, + "estimated_eval_tokens": 18672 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No true duplicates stand out in this set; most pairs are only loosely related by model loading, offloading, or tokenizer regressions. The broadest central issue is the offloading/quantization report, but the soft edges should all be rejected as distinct bugs.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43873 is the best umbrella issue here: it is broad, open, and sits at the center of the offloading/quantization-related soft matches.", + "summary": "This cluster is a set of largely unrelated bug reports across different models and subsystems. The soft-similarity links are mostly superficial (e.g., both mention loading, offload, device maps, or tokenization) but do not point to the same concrete bug or mergeable change set.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue: the items span different models, APIs, and failure modes, so picking one as the duplicate target would be misleading.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43873 is the most actionable representative of the cluster because it captures the recurring offload/quantization failure mode better than the narrower model-specific reports.", + "best_issue_reason": "No single best issue for deduplication; the cluster is heterogeneous, so none is a safe representative anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal indexing bugs, but they affect different models and different code paths (Qwen2.5-VL RoPE temporal positions vs Kosmos2.5 OCR long-input indexing)." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different models and failures: SigLIP2 produces bad results via AutoModel/pipeline, while Perceiver fails on non-default image resolution and interpolation." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both involve tied lm_head weights, but one is a model/config tie failure and the other is a serialization inconsistency by device; not the same concrete bug." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Different models and loading paths: Qwen3Moe FP8 configuration failure vs Qwen2.5-GGUF loading failure under v5." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "Unrelated failures: PaddleOCR-VL loading in vLLM versus a torch/transformers version incompatibility." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL RoPE/position-id scaling." + }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Different regressions: video-input generation flow vs Gemma4 device-mapping mismatch." + "reason": "Different regressions in different models: video-input handling for Qwen VL variants versus Gemma4 PLE device mismatch during forward." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Unrelated failures: init_empty_weights argument handling vs strict config loading for granite_speech." + "reason": "Different problems: init_empty_weights argument handling versus strict config rejection for granite_speech." }, { "left": "issue:43881", "right": "issue:44488", "accept": false, - "reason": "Both are load failures, but for different models and likely different code paths." + "reason": "Both are load failures, but for unrelated models and likely unrelated causes." }, { "left": "issue:43606", "right": "issue:44265", "accept": false, - "reason": "CPU offload device mismatch is separate from torch.export failing on torch_compilable_check." + "reason": "CPU offload device mismatch for Bark is unrelated to torch.export failure on torch_compilable_check models." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Offloading/quantization bug is unrelated to tokenizer changes across versions." + "reason": "Offloading/quantization behavior is unrelated to tokenization changes." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sampling-rate default is unrelated to PEFT checkpoint loading." + "reason": "Audio chat-template sampling-rate defaulting has no meaningful overlap with PEFT checkpoint loading." }, { - "left": "issue:34634", - "right": "issue:35707", + "left": "issue:40990", + "right": "issue:45584", "accept": false, - "reason": "Bark voice_preset bug and progressive generation with past_key_values are different subsystems." + "reason": "Different domains: perplexity on a language model versus Whisper generation failing on empty transcription." }, { - "left": "issue:40990", - "right": "issue:45584", + "left": "issue:34634", + "right": "issue:35707", "accept": false, - "reason": "Perplexity anomaly on gpt-oss is not the same bug as Whisper empty-transcription generation failure." + "reason": "Bark voice preset bug and progressive generation with inputs_embeds/past_key_values are separate issues." }, { "left": "issue:44568", "right": "issue:45920", "accept": false, - "reason": "Specific BOS/EOS omission for one tokenizer is not the same as broader wrong-token-id regressions." + "reason": "Both are tokenizer regressions, but they describe different symptoms and model families; no shared concrete bug is evident." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Inputs_embeds/past_key_values generation bug is unrelated to Qwen3 device-map cuda asserts." + "reason": "Progressive generation with inputs_embeds is unrelated to Qwen3 device-map CUDA assertion." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer padding/truncation crash is separate from export support for torch_compilable_check." + "reason": "Tokenizer NER/padding crashes and torch.export torch_compilable_check failures are separate code paths." }, { "left": "issue:43606", "right": "issue:43819", "accept": false, - "reason": "Both touch model behavior, but device offload mismatch is not the same as DAC STE mismatch." + "reason": "Both are audio-model-related, but one is offload device mismatch and the other is DAC from_latents vs forward mismatch due to missing STE." }, { "left": "issue:41720", "right": "issue:45823", "accept": false, - "reason": "Same broad area of device handling, but different models and different failure modes." - }, - { - "left": "issue:43454", - "right": "issue:43550", - "accept": false, - "reason": "Weight-tying bug for AyaVision is unrelated to Bamba torch.compile SDPA failure." - }, - { - "left": "issue:39692", - "right": "issue:40444", - "accept": false, - "reason": "Documentation/example errors are not the same as Qwen2.5-VL iterable-dataset finetuning failure." - }, - { - "left": "issue:43377", - "right": "issue:44610", - "accept": false, - "reason": "MIMI padding-mask mismatch is unrelated to OmDet processor image-size mismatch." - }, - { - "left": "issue:43454", - "right": "issue:45072", - "accept": false, - "reason": "AyaVision lm_head tying bug is distinct from dtype mismatch during bfloat16 inference." - }, - { - "left": "issue:43572", - "right": "issue:45759", - "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is not the same as text_config unwrapping for composite Qwen configs." - }, - { - "left": "issue:43317", - "right": "issue:43873", - "accept": false, - "reason": "Related offload theme, but one is dequantized-model loading and the other is a broader quantized offloading report." - }, - { - "left": "issue:43257", - "right": "issue:45923", - "accept": false, - "reason": "MOE weight conversion with accelerate+deepspeed is unrelated to a missing gradient-checkpointing flag." - }, - { - "left": "issue:43377", - "right": "issue:44898", - "accept": false, - "reason": "Padding-mask support in MIMI encoder is unrelated to Perceiver non-default-resolution failures." - }, - { - "left": "issue:43526", - "right": "issue:43883", - "accept": false, - "reason": "reduce_labels returning one label is unrelated to missing all_tied_weights_keys on Molmo." - }, - { - "left": "issue:40990", - "right": "issue:42947", - "accept": false, - "reason": "High perplexity on one model is not the same as ineffective gradient checkpointing with LoRA." + "reason": "Both mention device mapping, but the models and failure mechanisms differ (CUDA assert on Qwen3 vs Gemma4 PLE forward mismatch)." }, { - "left": "issue:43322", - "right": "issue:46050", + "left": "issue:43454", + "right": "issue:43550", "accept": false, - "reason": "Llava Next loading segfault is unrelated to sinq quantized save/load breakage." + "reason": "Tied-weight lm_head behavior and Bamba torch.compile SDPA failure are unrelated." }, { - "left": "issue:43881", - "right": "issue:44991", + "left": "issue:39692", + "right": "issue:40444", "accept": false, - "reason": "Different tokenizer/model loading failures with no shared concrete code path." + "reason": "Different multimodal workflows and different bugs: SigLIP2 docs/model mismatch versus Qwen2.5-VL IterableDataset multi-image finetuning failure." }, { - "left": "issue:43454", - "right": "issue:43720", + "left": "issue:43377", + "right": "issue:44610", "accept": false, - "reason": "lm_head weight tying and BitNet packed-weight unpacking are separate bugs." + "reason": "MIMI padding-mask batching bug and OmDet-Turbo processor/model input-size mismatch are unrelated." }, { - "left": "issue:41720", - "right": "issue:43322", + "left": "issue:43454", + "right": "issue:45072", "accept": false, - "reason": "Both are model-loading failures, but one is cuda assert under auto device mapping and the other is a segfault in Llava Next loading." + "reason": "Different issues around model behavior: tied lm_head weights versus dtype mismatches in bfloat16 inference." }, { - "left": "issue:43749", - "right": "issue:44534", + "left": "issue:43572", + "right": "issue:45759", "accept": false, - "reason": "FSDP CPU-efficient loading and non-persistent buffer junk are different regression classes." + "reason": "A missing pad_token_idx in StableLmConfig is unrelated to AutoModelForCausalLM.from_config not unwrapping text_config." }, { - "left": "issue:43296", - "right": "issue:44877", + "left": "issue:43257", + "right": "issue:45923", "accept": false, - "reason": "vLLM/PaddleOCR-VL loading failure is unrelated to granite_speech strict-config loading." + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed is unrelated to a missing supports_gradient_checkpointing flag." }, { - "left": "issue:44466", - "right": "issue:45357", + "left": "issue:43317", + "right": "issue:43873", "accept": false, - "reason": "Both involve serialization, but one is lm_head tying/device-dependent serialization and the other is visual-encoder key saving." + "reason": "Both involve offload/quantization, but one is loading dequantized models with device_map=auto and the other is a broader offloading expectation issue; not clearly the same bug." }, { - "left": "issue:41720", - "right": "issue:44977", + "left": "issue:43322", + "right": "issue:46050", "accept": false, - "reason": "Qwen3 device-map cudaErrorAssert is unrelated to flash-attention generation regression in Qwen3.5." + "reason": "Segmentation fault on Llava Next loading is unrelated to sinq quantized model save/load breakage." } ] }, @@ -3941,38 +3792,38 @@ "issue:42617", "issue:42915", "issue:43295", - "issue:43316", + "issue:43296", "issue:43317", + "issue:43322", "issue:43404", "issue:43450", + "issue:43454", "issue:43526", - "issue:43531", - "issue:43540", "issue:43550", - "issue:43572", "issue:43577", "issue:43606", "issue:43708", "issue:43720", - "issue:43782", + "issue:43749", "issue:43873", "issue:43881", "issue:44008", "issue:44186", "issue:44387", "issue:44423", + "issue:44466", "issue:44479", + "issue:44534", "issue:44610", "issue:44671", "issue:44743", "issue:44756", "issue:44877", "issue:44898", - "issue:44913", "issue:44918", + "issue:44977", "issue:44991", "issue:45072", - "issue:45081", "issue:45161", "issue:45357", "issue:45406", @@ -3980,249 +3831,263 @@ "issue:45684", "issue:45701", "issue:45704", - "issue:45812", "issue:45823" ], "soft_pairs": [ + "issue:43881|issue:44991", + "issue:43454|issue:43720", + "issue:41720|issue:43322", + "issue:43296|issue:44877", + "issue:43749|issue:44534", + "issue:44466|issue:45357", + "issue:41720|issue:44977", "issue:42617|issue:42915", "issue:40990|issue:45701", "issue:44387|issue:45704", "issue:44008|issue:45823", - "issue:44186|issue:45072", "issue:36331|issue:42617", + "issue:44186|issue:45072", "issue:43577|issue:43708", "issue:44877|issue:44991", - "issue:43404|issue:45357", "issue:36331|issue:42915", - "issue:43873|issue:44756", + "issue:43404|issue:45357", "issue:43450|issue:45701", + "issue:43873|issue:44756", "issue:43550|issue:43720", "issue:44423|issue:45406", "issue:41720|issue:42617", "issue:45632|issue:45684", "issue:44479|issue:44918", - "issue:44610|issue:44898", "issue:43550|issue:43606", + "issue:44610|issue:44898", "issue:44186|issue:44610", "issue:42915|issue:43317", "issue:43295|issue:45701", "issue:43526|issue:43881", "issue:41720|issue:45161", "issue:36010|issue:42617", - "issue:44671|issue:44743", - "issue:45081|issue:45812", - "issue:43531|issue:43572", - "issue:43316|issue:44913", - "issue:43540|issue:43782" + "issue:44671|issue:44743" ], - "prepared_review_unit_hash": "824869e8ad44a591819cd4e20ebd780e59b84f15fed3741e5a79ed01101b14c5", + "prepared_review_unit_hash": "b24425072f9083352865720ca9fa6c385311f89b6f8928a95f8b03e9c9687baa", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36857, - "estimated_input_tokens": 9215, - "estimated_eval_tokens": 18686 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 37126, + "estimated_input_tokens": 9282, + "estimated_eval_tokens": 18820 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is not a duplicate cluster: the items span many unrelated bugs across tokenization, multimodal processors, offloading, quantization, trainer behavior, and model-specific regressions. I found no pair that looks like the same underlying issue, so all soft edges are rejected.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "This cluster is mostly heterogeneous: the items cover unrelated loading, processor, quantization, multimodal, and trainer regressions. There is no strong duplicate group; at best, a few pairs share a broad subsystem, but not the same concrete bug.", + "confidence": 0.27, + "canonical_issue_reason": "issue:43873 is the broadest and most central open report in the set, covering offloading/quantization behavior that loosely overlaps several memory/offload-related items, so it is the best anchor if a canonical issue must be chosen.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "issue:43873 is the most generally framed and still active issue among the candidates, making it the least arbitrary representative of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Both are model-loading failures, but one is about a specific GLM vision model and the other about tokenizer loading for a different checkpoint; no shared code-path bug is evident." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Both mention weights during loading, but one is about tying lm_head weights and the other about unpacking packed BitNet weights; different mechanisms and fixes." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Both are inference/loading failures, but the reported symptoms and model paths differ (Qwen3 auto device mapping vs Llava Next segfault)." + }, + { + "left": "issue:43296", + "right": "issue:44877", + "accept": false, + "reason": "These are both load failures, but one is a vLLM/transformers compatibility issue and the other is strict config validation; different root causes." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "Both touch memory/offload behavior, but one is an FSDP CPU RAM-efficient loading regression and the other is a v5 non-persistent buffer serialization problem." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both involve saving/serialization regressions, but one is tied weights serialization and the other is incorrect visual encoder key saving; not the same bug." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Both involve Qwen3.5-family models, but one is a device-mapping crash and the other is a flash-attention generation regression." + }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "Different failure modes: 3d_parallel.py runtime failure vs Qwen3Moe FP8Config loading issue." + "reason": "Unrelated failures: a 3d_parallel.py runtime problem versus a Qwen3Moe FineGrainedFP8Config issue." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "Different symptoms and scope: high perplexity on a specific model vs a general tokenization-version regression." + "reason": "Tokenizer/tokenization behavior could affect perplexity, but the reports do not describe the same underlying defect or code path." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Both mention memory, but one is int4 quantization OOM and the other is an Apex RMSNorm leak in T5." + "reason": "Both are memory-related, but one is int4 quantization causing extra reserved CUDA memory and the other is an apex RMSNorm memory leak." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Different models and code paths: Gemma3n forward variable collision vs Gemma4 PLE device mismatch." + "reason": "Both are Gemma-family model bugs, but one is a variable-name collision in forward and the other is a device-mapping mismatch in a different model." }, { - "left": "issue:44186", - "right": "issue:45072", + "left": "issue:36331", + "right": "issue:42617", "accept": false, - "reason": "Tokenizer crashes on NER/padding vs dtype mismatches in inference; unrelated bugs." + "reason": "A trainer API signature change is unrelated to the 3d_parallel.py failure." }, { - "left": "issue:36331", - "right": "issue:42617", + "left": "issue:44186", + "right": "issue:45072", "accept": false, - "reason": "CustomTrainer loss-signature regression is unrelated to 3d_parallel.py execution failure." + "reason": "Different subsystems entirely: tokenizer padding/truncation versus bfloat16 dtype mismatches in model inference." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "BLIP2 dtype loading bug vs Trainer checkpoint resume step-count bug; different layers of the stack." + "reason": "One is a dtype-loading bug in BLIP2, the other is a resume/max_steps calculation bug in Trainer; no shared code path." }, { "left": "issue:44877", "right": "issue:44991", "accept": false, - "reason": "Strict config loading and tokenizer loading failures are different issues with different root causes." - }, - { - "left": "issue:43404", - "right": "issue:45357", - "accept": false, - "reason": "One is weight tying in Mistral3, the other is save_pretrained emitting wrong visual encoder keys for Qwen3.5." + "reason": "Both are loading regressions, but one is strict config handling and the other is tokenizer loading for a specific model; not the same defect." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Trainer API signature breakage is unrelated to Qwen3Moe FP8 loading failure." + "reason": "Trainer loss-signature handling and FP8 model loading are unrelated." }, { - "left": "issue:43873", - "right": "issue:44756", + "left": "issue:43404", + "right": "issue:45357", "accept": false, - "reason": "Both involve memory/offload, but one is quantization offloading behavior and the other is mmap OOM on Strix Halo." + "reason": "Both concern multimodal model internals, but one is weight tying and the other is save_pretrained key serialization; different bugs." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Video batching shape bug vs tokenization changes; no shared underlying defect." + "reason": "Video processor batching shape issues are unrelated to the tokenization regression claim." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Both are memory/offload themed, but one is quantization offloading and the other is mmap-related OOM avoidance; distinct root causes." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "torch.compile with SDPA on Bamba vs packed-weight unpacking during accelerate loading for BitNet; different code paths." + "reason": "One is a torch.compile/SDPA failure in Bamba, the other is an accelerate loading issue for BitNet packed weights." }, { "left": "issue:44423", "right": "issue:45406", "accept": false, - "reason": "Both affect serve, but one is continuous-batching multimodal crash and the other is a missing _tokenizer attribute on Gemma4Processor." - }, - { - "left": "issue:41720", - "right": "issue:42617", - "accept": false, - "reason": "Qwen3 auto device mapping cuda assert is unrelated to the 3d_parallel.py failure." + "reason": "Both are serve-time multimodal failures, but the specific crashes differ and point to different processor/code-path issues." }, { "left": "issue:45632", "right": "issue:45684", "accept": false, - "reason": "Cache-path collision for trust_remote_code is unrelated to save_pretrained permission propagation." + "reason": "Cache-path collision and read-only permission propagation are unrelated save/load concerns." }, { "left": "issue:44479", "right": "issue:44918", "accept": false, - "reason": "Video-input regression across Qwen VL models vs unpacking Qwen3.5 embeddings in TRL SFT trainer; not the same bug." + "reason": "Both mention Qwen3.5, but one is a video-input regression and the other is TRL SFT embedding unpacking; not the same bug." }, { - "left": "issue:44610", - "right": "issue:44898", + "left": "issue:43550", + "right": "issue:43606", "accept": false, - "reason": "Both are image-size/preprocessing related, but they affect different models and failure conditions." + "reason": "Different failures: torch.compile with SDPA versus CPU offload device mismatch." }, { - "left": "issue:43550", - "right": "issue:43606", + "left": "issue:44610", + "right": "issue:44898", "accept": false, - "reason": "Different models and triggers: Bamba SDPA/compile issue vs Bark CPU offload device mismatch." + "reason": "Both are image-size/preprocessing issues, but they involve different models and different expected-input mismatches." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Tokenizer padding/truncation crash and OmDet-Turbo size mismatch are separate preprocessing bugs." + "reason": "Tokenizer padding/NER crashes and image processor resolution mismatch are different preprocessing bugs." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Qwen3Moe FP8 loading failure is not the same as dequantized model offload loading failure." + "reason": "FP8 model failure and dequantized-model offload failure are related only at a very high level; they are not the same code-path problem." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Processor/tokenizer API regression is distinct from a general tokenization version-change report." + "reason": "A regression around processor.tokenizer/image handling is not the same as a general tokenization-change report." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "BEiT label reduction bug and glm-4v-9b loading failure are unrelated." + "reason": "Different components and symptoms: BEiT label reduction versus GLM model loading." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Auto device map cuda assert and GPT-OSS tensor-parallel-only failure are different model/runtime problems." + "reason": "Tensor-parallel GPT-OSS MoE issues and Qwen3 auto device mapping failures are different model/runtime problems." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to 3d_parallel.py execution problems." + "reason": "Importing GenerationMixin is unrelated to a 3d_parallel.py execution failure." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "CamemBERT masked-LM regression and Qwen3.5 recurrent-state cache bug are unrelated model-specific defects." - }, - { - "left": "issue:45081", - "right": "issue:45812", - "accept": false, - "reason": "Mistral regex patch crash and Granite token ID regression are both tokenizer issues but have different causes and code paths." - }, - { - "left": "issue:43531", - "right": "issue:43572", - "accept": false, - "reason": "Qwen3-MoE sliding-window behavior and StableLmConfig pad_token_idx omission are unrelated." - }, - { - "left": "issue:43316", - "right": "issue:44913", - "accept": false, - "reason": "Gemma3TextConfig API mismatch is unrelated to GPTNeoX rotary_pct reloading behavior." - }, - { - "left": "issue:43540", - "right": "issue:43782", - "accept": false, - "reason": "Qwen3OmniMoe video-processing ValueError and Qwen3VL weight_only loading error are different failures." + "reason": "CamemBERT masked-LM prediction errors and Qwen3.5 recurrent-state cache resets are unrelated model-specific bugs." } ] }, @@ -4257,12 +4122,15 @@ "issue:41720", "issue:42490", "issue:42915", + "issue:43316", "issue:43322", "issue:43329", "issue:43388", "issue:43421", "issue:43531", + "issue:43540", "issue:43550", + "issue:43572", "issue:43606", "issue:43653", "issue:43720", @@ -4270,7 +4138,6 @@ "issue:43782", "issue:43873", "issue:43927", - "issue:44077", "issue:44186", "issue:44361", "issue:44451", @@ -4278,23 +4145,26 @@ "issue:44509", "issue:44514", "issue:44545", - "issue:44589", "issue:44625", "issue:44898", + "issue:44913", "issue:44991", "issue:45059", "issue:45072", - "issue:45381", + "issue:45081", "issue:45406", "issue:45464", "issue:45479", "issue:45701", "issue:45704", "issue:45759", - "issue:45812", - "issue:45859" + "issue:45812" ], "soft_pairs": [ + "issue:45081|issue:45812", + "issue:43531|issue:43572", + "issue:43316|issue:44913", + "issue:43540|issue:43782", "issue:42915|issue:43322", "issue:44186|issue:44361", "issue:43421|issue:45812", @@ -4302,237 +4172,233 @@ "issue:29159|issue:36032", "issue:43388|issue:45059", "issue:45406|issue:45464", - "issue:30064|issue:34634", "issue:33357|issue:35707", + "issue:30064|issue:34634", "issue:43606|issue:43720", - "issue:41720|issue:43746", "issue:43550|issue:43653", + "issue:41720|issue:43746", "issue:43873|issue:45704", "issue:43421|issue:45701", "issue:44492|issue:44509", "issue:43606|issue:43653", - "issue:34634|issue:41720", "issue:43531|issue:45759", + "issue:34634|issue:41720", "issue:42490|issue:43329", "issue:34567|issue:35707", "issue:44514|issue:44545", "issue:44625|issue:45479", "issue:43421|issue:43927", "issue:44898|issue:45072", - "issue:44451|issue:44991", - "issue:45759|issue:45859", - "issue:44077|issue:44625", - "issue:43531|issue:44589", - "issue:43782|issue:45381" + "issue:44451|issue:44991" ], - "prepared_review_unit_hash": "13929a02926103434cc986a7f8ee8295f5774c9146a287e2c9d0403893b8cdda", + "prepared_review_unit_hash": "02752ac3460fa627f400bc3317b32f9cf468b4c3d7e1138853795ac26a2485fe", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36004, - "estimated_input_tokens": 9001, - "estimated_eval_tokens": 18258 + "serialized_chars": 36630, + "estimated_input_tokens": 9158, + "estimated_eval_tokens": 18572 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issue set; only #44514 and #44545 are the same crash. The rest are false-positive similarity matches across different models/subsystems.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue fits the whole cluster because the items are largely unrelated. If a local representative is needed, #44514 is the clearest choice because #44545 is essentially the same bug with the same title.", + "summary": "This set is mostly heterogeneous; the only clear duplicate is the Qwen2_5_VLProcessor batched `padding=False` crash (44514/44545). All other soft pairs share a subsystem at most, but not the same concrete bug.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 44514 is the cleaner anchor for the only true duplicate pair here: it matches 44545 exactly on the same `Qwen2_5_VLProcessor.apply_chat_template` batched `padding=False` crash.", "canonical_pr_reason": null, - "best_issue_reason": "#44514 is the best issue representative for the only true duplicate pair (#44545): same API, same padding=False batched-input crash, clearer wording.", + "best_issue_reason": "44514 is the most actionable and clearly duplicated issue in the set; it best represents the one confirmed bug cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer-related, but they describe different failures: Mistral regex patching crash vs Granite token ID regression." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Different model/config bugs: Qwen3-MoE sliding window handling vs StableLmConfig missing `pad_token_idx`." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Both are config discrepancies, but for different fields and models (`Gemma3TextConfig` API vs GPTNeoX `rotary_pct` reload behavior)." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Different code paths: Qwen3OmniMoe video input processing vs Qwen3VL `from_pretrained` weight-only loading." + }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "Different models and failure modes: FineGrainedFP8Config vs Llava Next load-time segfault." + "reason": "Different failures and model families: FineGrainedFP8Config issue vs Llava Next segmentation fault on load." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Both tokenizer-ish, but one is LayoutLMv2 NER/batching crash and the other is MLuke task AttributeError; not the same bug." + "reason": "Both hit tokenizers, but the bugs are different: NER/batched padding crash vs MLuke task-time AttributeError." }, { "left": "issue:43421", "right": "issue:45812", "accept": false, - "reason": "Both involve tokenizers, but one is runtime post-processor updates and the other is wrong token IDs for Granite models; different code paths." + "reason": "Runtime special-token post-processor sync bug is not the same as Granite tokenizer returning wrong token IDs." }, { "left": "issue:36331", "right": "issue:41720", "accept": false, - "reason": "Trainer signature mismatch vs Qwen3 auto device-mapping CUDA assert; unrelated." + "reason": "Trainer signature mismatch and Qwen3 device-mapping CUDA assert are unrelated." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "Slow/fast tokenizer inconsistency is not the same as T5 tokenizer load-time add_special_tokens conflict." + "reason": "Both involve tokenizers, but one is slow/fast inconsistency and the other is a T5 special-token name conflict on load." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "Metric gathering label truncation is unrelated to SAM3 text/box behavior." + "reason": "Metric-gathering label truncation is unrelated to SAM3 text/bounding-box processing." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Gemma4Processor missing _tokenizer and Qwen3.5 streaming API failure are different processor/runtime issues." + "reason": "Different surfaces and failures: missing `_tokenizer` in `Gemma4Processor` vs Qwen3.5 streaming chat/completions failure." }, { - "left": "issue:30064", - "right": "issue:34634", + "left": "issue:33357", + "right": "issue:35707", "accept": false, - "reason": "Image processor void-segmentation-map handling is unrelated to BarkProcessor voice_preset support." + "reason": "A MacOS CLIP bus error is not the same bug as progressive generation with `inputs_embeds`/`past_key_values`." }, { - "left": "issue:33357", - "right": "issue:35707", + "left": "issue:30064", + "right": "issue:34634", "accept": false, - "reason": "MacOS CLIP bus error and progressive generation with inputs_embeds/past_key_values are not the same failure." + "reason": "Image processor segmentation-map handling and Bark voice_preset behavior are unrelated." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "CPU offload device mismatch vs packed-weight unpacking during accelerate loading; different bugs." + "reason": "Both touch loading/offload paths, but one is CPU-offload device mismatch and the other is BitNet packed-weight unpacking." }, { - "left": "issue:41720", - "right": "issue:43746", + "left": "issue:43550", + "right": "issue:43653", "accept": false, - "reason": "Qwen3 auto device mapping CUDA assert does not match GraniteSpeech PEFT local-checkpoint loading." + "reason": "Different model/paths: torch.compile + SDPA failure vs tokenizer special-token registration bug." }, { - "left": "issue:43550", - "right": "issue:43653", + "left": "issue:41720", + "right": "issue:43746", "accept": false, - "reason": "torch.compile/SDPA issue is unrelated to BigBirdTokenizer mask token registration." + "reason": "CUDA assert during auto device mapping is unrelated to GraniteSpeech PEFT local-checkpoint loading." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantized offloading issue is unrelated to T5 apex FusedRMSNorm memory leak." + "reason": "Quantized offloading behavior and T5 Apex RMSNorm memory leak are separate issues." }, { "left": "issue:43421", "right": "issue:45701", "accept": false, - "reason": "Runtime special-token/post-processor updates are not the same as version-dependent tokenization drift." + "reason": "Special-token post-processor refresh is not the same as a general tokenization regression across versions." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "Doc typo vs removed pipeline-task docs; different documentation issues, not duplicates." + "reason": "A cache-strategy typo and stale docs about removed pipeline tasks are unrelated." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Bark CPU-offload device mismatch is not the same as BigBirdTokenizer decode/output special-token bug." + "reason": "Same project area, but different concrete bugs: CPU-offload mismatch vs BigBird tokenizer mask token registration." }, { - "left": "issue:34634", - "right": "issue:41720", + "left": "issue:43531", + "right": "issue:45759", "accept": false, - "reason": "Bark voice_preset problem and Qwen3 auto device mapping CUDA assert are unrelated." + "reason": "Both involve Qwen3.5/Qwen3-MoE config handling, but one is sliding-window behavior and the other is `from_config` text_config unwrapping." }, { - "left": "issue:43531", - "right": "issue:45759", + "left": "issue:34634", + "right": "issue:41720", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to unwrapping text_config in composite Qwen configs." + "reason": "BarkProcessor voice preset and Qwen3 auto device mapping are unrelated." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "Trailing-slash local model import warning is unrelated to multimodal video-token helper undefined names." + "reason": "Import-path slash warning is unrelated to undefined video-token helper variables in multimodal processors." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token counter update bug and progressive generation bug are different components." + "reason": "Trainer token-count tracking and progressive generation are different bugs." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "Same Qwen2_5_VLProcessor.apply_chat_template crash on batched input with padding=False; titles and behavior match closely." + "reason": "Same exact bug report and same crash condition: `Qwen2_5_VLProcessor.apply_chat_template` on batched input with `padding=False`." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Num-label propagation bug is distinct from zero-loss behavior when problem_type=single_label_classification and num_labels=1." + "reason": "Both concern classification labels, but one is config propagation and the other is a degenerate loss formulation with `num_labels=1`." }, { "left": "issue:43421", "right": "issue:43927", "accept": false, - "reason": "Special-token post-processor update issue is unrelated to DiaConfig losing custom token IDs on save/load." + "reason": "Runtime special-token post-processor updates are a different bug from DiaConfig losing custom token IDs after save/load." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver interpolation failure and dtype mismatch in SwitchTransformers/TimmWrapperModel are different inference bugs." + "reason": "Image-resolution interpolation failure is unrelated to dtype mismatch in bf16 inference." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Generic model-loading failure and tokenizer regression for EMBEDDIA/est-roberta are not the same issue." - }, - { - "left": "issue:45759", - "right": "issue:45859", - "accept": false, - "reason": "Composite config unwrapping and Qwen3_5MoeTextRotaryEmbedding CPU-offload compatibility are different code paths." - }, - { - "left": "issue:44077", - "right": "issue:44625", - "accept": false, - "reason": "post_init allowance in patchtsmixer is unrelated to num_labels propagation in Qwen3.5." - }, - { - "left": "issue:43531", - "right": "issue:44589", - "accept": false, - "reason": "Qwen3-MoE sliding_window and missing Float8 storage type are unrelated." - }, - { - "left": "issue:43782", - "right": "issue:45381", - "accept": false, - "reason": "Qwen3VL weight_only load error and Qwen2.5-VL video vision_position_ids bug are different model/version issues." + "reason": "Both are tokenizer-loading regressions, but they affect different models and do not look like the same underlying defect." } ] }, "evaluator_result": { "accept": true, - "feedback": "Accepted. The cluster summary is grounded in the packet, and the soft-edge verdicts stay conservative. The one accepted duplicate pair (#44514/#44545) is well supported by the matching title and same batched `padding=False` crash; the rest are reasonably rejected as different bugs/subsystems." + "feedback": "The summary is well grounded in the packet: the 44514/44545 pair is an exact title-level duplicate with the same `Qwen2_5_VLProcessor.apply_chat_template` batched `padding=False` crash, and the other soft pairs are handled conservatively without overstating evidence. No mergeability issues are apparent." }, "evaluator_used": true, "retried": false, @@ -4562,25 +4428,25 @@ "issue:43381", "issue:43421", "issue:43531", - "issue:43540", "issue:43596", - "issue:43618", "issue:43645", "issue:43653", - "issue:43697", "issue:43720", "issue:43746", + "issue:43782", "issue:43819", "issue:43844", "issue:43873", "issue:43994", - "issue:44008", "issue:44062", + "issue:44077", "issue:44186", "issue:44220", "issue:44368", "issue:44451", + "issue:44589", "issue:44610", + "issue:44625", "issue:44661", "issue:44671", "issue:44843", @@ -4589,10 +4455,9 @@ "issue:44987", "issue:45042", "issue:45081", - "issue:45183", "issue:45356", "issue:45357", - "issue:45479", + "issue:45381", "issue:45588", "issue:45663", "issue:45704", @@ -4603,22 +4468,26 @@ "issue:46050" ], "soft_pairs": [ + "issue:45759|issue:45859", + "issue:44077|issue:44625", + "issue:43531|issue:44589", + "issue:43782|issue:45381", "issue:44987|issue:45042", "issue:45588|issue:45759", - "issue:44368|issue:45357", "issue:44062|issue:44451", + "issue:44368|issue:45357", "issue:39692|issue:44610", "issue:43746|issue:46050", "issue:45704|issue:45923", "issue:44918|issue:44936", - "issue:45663|issue:45923", "issue:43421|issue:43653", + "issue:45663|issue:45923", "issue:43381|issue:43844", "issue:42886|issue:44843", "issue:45704|issue:45859", "issue:43531|issue:43645", - "issue:43421|issue:44186", "issue:41720|issue:45704", + "issue:43421|issue:44186", "issue:43994|issue:44671", "issue:43746|issue:43873", "issue:39692|issue:43873", @@ -4629,226 +4498,232 @@ "issue:30064|issue:36010", "issue:43720|issue:43819", "issue:44220|issue:44661", - "issue:43653|issue:43720", - "issue:43540|issue:44008", - "issue:43618|issue:43697", - "issue:45183|issue:45479" + "issue:43653|issue:43720" ], - "prepared_review_unit_hash": "44702792a48e7e9c21d60818ce3ec1b0acb397b61100d548f5f63bcb36fcf56f", + "prepared_review_unit_hash": "4a9c0e53199f851dce4bf3c287d9888061b53b55ea916e3e47ceb13063eb368d", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 37215, - "estimated_input_tokens": 9304, - "estimated_eval_tokens": 18864 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36774, + "estimated_input_tokens": 9194, + "estimated_eval_tokens": 18644 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: it mixes unrelated tokenizer, multimodal, training, loading, and documentation regressions. None of the soft pairs look like the same underlying bug or change, so all should stay separate.", - "confidence": 0.96, - "canonical_issue_reason": "No clear canonical issue for this cluster: the items span many unrelated defects and do not converge on one underlying bug.", + "summary": "This cluster is mostly heterogeneous and only one soft edge looks like a true duplicate: the offline tokenizer/cache regression around HF_HUB_OFFLINE. The rest are distinct bugs across unrelated model, tokenizer, training, and quantization paths.", + "confidence": 0.31, + "canonical_issue_reason": "Issue 44843 is the clearest root-cause-style report in the cluster and appears to subsume the offline tokenizer/cache failure described in 42886 via the _patch_mistral_regex/model_info path.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good global representative here; any candidate would be misleading because the cluster is not a true duplicate set.", + "best_issue_reason": "Issue 44843 is the strongest representative because it isolates a specific failing code path rather than only a symptom, but the overall cluster is not a single coherent duplicate set.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Different bugs: composite-config unwrapping in from_config vs CPU-offload incompatibility in rotary embedding." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "Unrelated model/config issues; one is a patchtsmixer post_init policy change, the other is Qwen3.5 num_labels propagation." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Different subsystems and failures: Qwen3-MoE sliding window vs missing Float8 storage type." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen-VL related, but one is weight_only loading and the other is video vision_position_ids; not the same bug." + }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Different failures: one is about loading a specific model under transformers>=5.1.0, the other is a PIL image-processor/torchvision dependency regression." + "reason": "Different issues: model loading regression for one package vs PIL image processors wrongly requiring torchvision." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Unrelated code paths: flash-attention crashes on sink-less models, while the other is about unwrapping text_config in composite Qwen multimodal configs." + "reason": "Flash-attention sink-less model crash is unrelated to Qwen composite config unwrapping." }, { - "left": "issue:44368", - "right": "issue:45357", + "left": "issue:44062", + "right": "issue:44451", "accept": false, - "reason": "Both mention Qwen3.5, but one is a LoRA warning about tie_word_embeddings and the other is a save_pretrained regression for visual encoder keys." + "reason": "Different failure modes: AddedToken special-arg TypeError vs a model load failure for ScandiBERT." }, { - "left": "issue:44062", - "right": "issue:44451", + "left": "issue:44368", + "right": "issue:45357", "accept": false, - "reason": "A tokenizer AddedToken constructor error is not the same bug as failing to load a particular model." + "reason": "One is a tie_word_embeddings warning, the other is incorrect visual-encoder key saving; not the same code path." }, { "left": "issue:39692", "right": "issue:44610", "accept": false, - "reason": "Both involve model/processor mismatches, but they are different models and different root causes (docs example/quantization vs image-size mismatch)." + "reason": "Both mention vision models, but one is a SigLIP2 doc/example mismatch and the other is an OmDet-Turbo processor size bug." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "PEFT adapter local-checkpoint loading and sinq quantized save/load breakage are distinct failure modes." + "reason": "Different checkpoint/loading problems: PEFT adapter local load vs sinq quantized save/load corruption." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "T5 apex RMSNorm memory leak is unrelated to a missing supports_gradient_checkpointing flag in a trust_remote_code model." + "reason": "T5 RMSNorm memory leak is unrelated to Nemotron gradient-checkpointing flag support." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "Different trainer problems: unpacking embeddings in SFT versus evaluate() failing after train()." + "reason": "Different training failures: Qwen3.5 embedding unpacking in TRL SFT vs trainer.evaluate() after trainer.train()." }, { - "left": "issue:45663", - "right": "issue:45923", + "left": "issue:43421", + "right": "issue:43653", "accept": false, - "reason": "FSDP2 shared_kv_states rebuilding is a different bug from the missing gradient-checkpointing flag." + "reason": "Related to special tokens, but not the same bug: runtime post-processor updates vs BigBirdTokenizer mask-token registration/decode output." }, { - "left": "issue:43421", - "right": "issue:43653", + "left": "issue:45663", + "right": "issue:45923", "accept": false, - "reason": "One is a runtime feature gap in TokenizersBackend post-processing; the other is a BigBirdTokenizer special-token registration bug." + "reason": "Gemma-4 FSDP2 shared_kv_states KeyError is unrelated to gradient-checkpointing flag metadata." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Eval-mode checkpointing is a different issue from gradients abnormally increasing under ZeRO-3." + "reason": "Eval-mode gradient checkpointing limitation is unrelated to a ZeRO-3 gradient blow-up." }, { "left": "issue:42886", "right": "issue:44843", - "accept": false, - "reason": "Both are offline-tokenizer failures, but 44843 is a specific model_info() call in _patch_mistral_regex while 42886 is a broader cache-loading report; not enough to treat as the same bug." + "accept": true, + "reason": "Same underlying offline tokenizer/cache regression; 44843 identifies the culprit (_patch_mistral_regex calling model_info unconditionally) behind 42886's HF_HUB_OFFLINE failure." }, { "left": "issue:45704", "right": "issue:45859", "accept": false, - "reason": "RMSNorm memory leak in T5 and CPU-offload incompatibility in Qwen3_5MoeTextRotaryEmbedding are unrelated." + "reason": "Different issues: T5 RMSNorm memory leak vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility." }, { "left": "issue:43531", "right": "issue:43645", "accept": false, - "reason": "A sliding_window issue in Qwen3-MoE is unrelated to notebook-defined custom model initialization breaking in Transformers 5.0.0." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to Jupyter custom-model initialization in Transformers 5.0.0." }, { - "left": "issue:43421", - "right": "issue:44186", + "left": "issue:41720", + "right": "issue:45704", "accept": false, - "reason": "Runtime special-token updates and LayoutLMv2Tokenizer NER/padding crashes are different tokenizer bugs." + "reason": "CUDA assert from auto device mapping is unrelated to the T5 apex RMSNorm memory leak." }, { - "left": "issue:41720", - "right": "issue:45704", + "left": "issue:43421", + "right": "issue:44186", "accept": false, - "reason": "Qwen3 auto device-mapping cuda assert is unrelated to the T5 RMSNorm memory leak." + "reason": "Special-token post-processor behavior is unrelated to LayoutLMv2Tokenizer NER/padding crashes." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "SigLIP2 inference/processor issues are not the same as a CamemBERT masked-LM regression." + "reason": "Different model families and bugs: SigLIP2 inference regression vs CamemBERT masked-LM prediction regression." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "Local checkpoint loading with PEFT adapters and offloading behavior with quantization are distinct problems." + "reason": "Local checkpoint loading with PEFT adapters is unrelated to quantization offloading behavior." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "Docs/model mismatch for SigLIP2 is not the same as offloading not working as expected with quantization." + "reason": "Doc/example errors for SigLIP2 are unrelated to quantization/offloading behavior." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Both mention tokenizer regression and fix_mistral_regex, but one is a crash in _patch_mistral_regex for Mistral tokenizers and the other is a Kimi codec-handling regression; not the same concrete bug." + "reason": "Both involve tokenizer regex fixes, but one is a hard crash loading Mistral tokens and the other is a codec-handling regression in Kimi-K2.5; not clearly the same bug." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "DeepSpeed ZeRO-3/BertModel init IndexError and _torch_extract_fbank_features() are unrelated." + "reason": "Different low-level runtime errors: DeepSpeed ZeRO-3/BertModel index error vs _torch_extract_fbank_features issue." }, { "left": "issue:40444", "right": "issue:43746", "accept": false, - "reason": "Qwen2.5-VL multi-image IterableDataset finetuning failure is unrelated to GraniteSpeech local-checkpoint loading with PEFT adapters." + "reason": "Multimodal iterable dataset fine-tuning failure is unrelated to PEFT local checkpoint loading." }, { "left": "issue:45356", "right": "issue:45907", "accept": false, - "reason": "Tokenizer codec handling/regression has nothing to do with list_repo_files failing to forward revision in CI." + "reason": "Tokenizer codec/regex regression is unrelated to list_repo_files not forwarding revision." }, { "left": "issue:30064", "right": "issue:36010", "accept": false, - "reason": "Image-processor void segmentation-map handling and a GenerationMixin import error are completely different issues." + "reason": "Completely different topics: void segmentation map processing vs GenerationMixin import error." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "Packed-weight loading in BitNet and DAC.from_latents vs forward mismatch are different model-specific bugs." + "reason": "Different numerical/model bugs: packed-weight loading vs DAC.from_latents missing STE mismatch." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "An audio fbank feature-extraction problem is unrelated to add-new-model-like failing inside TOKENIZER_MAPPING_NAMES." + "reason": "Audio feature extraction bug is unrelated to add-new-model-like failing for TOKENIZER_MAPPING_NAMES." }, { "left": "issue:43653", "right": "issue:43720", "accept": false, - "reason": "BigBirdTokenizer special-token decode output and BitNet packed-weight unpacking are unrelated." - }, - { - "left": "issue:43540", - "right": "issue:44008", - "accept": false, - "reason": "Qwen3OmniMoe video-input processing and Gemma 3n audio_mel_mask variable-collision bugs are different code paths." - }, - { - "left": "issue:43618", - "right": "issue:43697", - "accept": false, - "reason": "Missing CLIPOutput attentions assignment and RTDetrV2 output drift are separate regressions." - }, - { - "left": "issue:45183", - "right": "issue:45479", - "accept": false, - "reason": "Validation error messaging for input_ids/inputs_embeds is not the same bug as degenerate zero loss in single-label classification." + "reason": "BigBird special-token decode issue is unrelated to BitNet packed-weight loading." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only likely duplicate edge called out (42886 \u2194 44843) is supported by the issue titles and the shared _patch_mistral_regex/model_info path, while the rest of the soft-edge rejections are plausibly distinct bugs. No clear overstatement beyond the cautious wording already used." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -4878,11 +4753,11 @@ "issue:43540", "issue:43550", "issue:43577", + "issue:43618", "issue:43643", - "issue:43644", "issue:43653", + "issue:43697", "issue:43716", - "issue:43749", "issue:43756", "issue:43782", "issue:43818", @@ -4894,7 +4769,6 @@ "issue:44442", "issue:44448", "issue:44488", - "issue:44561", "issue:44671", "issue:44811", "issue:44821", @@ -4903,11 +4777,12 @@ "issue:44898", "issue:44977", "issue:44991", - "issue:45020", "issue:45072", "issue:45081", "issue:45084", + "issue:45183", "issue:45381", + "issue:45479", "issue:45584", "issue:45685", "issue:45834", @@ -4915,7 +4790,10 @@ "issue:46032" ], "soft_pairs": [ + "issue:43540|issue:44008", + "issue:43618|issue:43697", "issue:43550|issue:45072", + "issue:45183|issue:45479", "issue:43295|issue:45834", "issue:43550|issue:43819", "issue:43388|issue:46032", @@ -4926,213 +4804,217 @@ "issue:44186|issue:44898", "issue:43653|issue:43819", "issue:43531|issue:43643", - "issue:43950|issue:45987", "issue:43643|issue:44877", + "issue:43950|issue:45987", "issue:44843|issue:45081", "issue:30064|issue:36331", "issue:43540|issue:45381", "issue:44448|issue:44671", "issue:43421|issue:44811", "issue:41720|issue:44991", - "issue:43577|issue:43818", "issue:39692|issue:45834", + "issue:43577|issue:43818", "issue:43782|issue:44977", - "issue:44821|issue:45084", "issue:45584|issue:45834", - "issue:44186|issue:44442", - "issue:43644|issue:43749", - "issue:44561|issue:45020" + "issue:44821|issue:45084", + "issue:44186|issue:44442" ], - "prepared_review_unit_hash": "b5de206bfda36d964a6e8b8d8cc7492d06f92c6e6a4ec6c459da788af5097a25", + "prepared_review_unit_hash": "53ec5b1d296865e1d3fa27dd7e381108dd19a9c322ce8055d21c4186ff074185", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 36049, - "estimated_input_tokens": 9013, - "estimated_eval_tokens": 18282 + "soft_pair_count": 28, + "serialized_chars": 36507, + "estimated_input_tokens": 9127, + "estimated_eval_tokens": 18510 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are distinct model- or subsystem-specific bugs. None of the soft pairs looks like a true duplicate; the closest pair (the two Mistral regex reports) still describe different failure modes in the same helper.", - "confidence": 0.84, - "canonical_issue_reason": "issue:45020 is the broadest, most umbrella-like report in the set: it frames a general regression affecting recent transformers versions and remote_code models, making it the best canonical anchor for that subgroup.", + "summary": "This cluster is heterogeneous: the soft pairs share broad Transformers themes but not the same concrete bug or fix path, so I would not deduplicate any of them.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out because the cluster mixes unrelated regressions across models, tokenizers, processors, cache logic, and training utilities; if forced to pick one anchor, issue 44843 is the clearest core-loading regression, but it does not represent the rest.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the most suitable global issue because it is the most general, clearly stated regression among otherwise narrow, model-specific reports.", + "best_issue_reason": "Issue 44843 is the most concrete and broadly actionable regression in the set, with a clear failure mode and code path, though it is not a true representative of the whole cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different models and root causes: Qwen3OmniMoe video-processing ValueError vs Gemma 3n audio_mel_mask attribute collision." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Both are model output regressions, but one is missing CLIP attentions and the other is RTDetrV2 producing different outputs; not the same bug." + }, { "left": "issue:43550", "right": "issue:45072", "accept": false, - "reason": "Different models and failures: torch.compile/SDPA vs bfloat16 dtype mismatch." + "reason": "Distinct code paths and failures: Bamba torch.compile/SDPA issue vs SwitchTransformers/TimmWrapperModel dtype mismatch." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "One is input validation/error-message logic; the other is zero-loss behavior in single-label classification. Different bugs." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Different multimodal pipelines and errors; no shared concrete code path." + "reason": "Both involve model loading/processing, but the regression in processor.tokenizer access is unrelated to Kosmos2.5's long-OCR index error." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Unrelated bugs in different model implementations." + "reason": "Unrelated failures: torch.compile with SDPA in Bamba vs DAC.from_latents not matching forward pass due to missing STE." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Different runtime paths: metric-gathering label truncation vs cache/seq_len correctness." + "reason": "Different areas: gather_for_metrics last-batch label handling vs Mamba2Mixer cache correctness for seq_len > 1." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "Both tokenizer-related, but one is an AddedToken argument bug and the other is a model loading failure." + "reason": "Tokenizer construction bug with AddedToken kwargs vs failure to load a specific model tokenizer; no shared root cause." }, { "left": "issue:43716", "right": "issue:44008", "accept": false, - "reason": "Different components: image preprocessor dtype mismatch vs forward-pass variable collision." + "reason": "Different models and modalities: Blip2 image preprocessor dtype mismatch vs Gemma 3n variable-name collision in audio forward pass." }, { "left": "issue:43335", "right": "issue:43756", "accept": false, - "reason": "Different model families and unrelated architectural bugs." + "reason": "Different model/config bugs: sparse layer creation with zero layers vs Smollm3 RoPE layer count mismatch." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "A segfault on Llava Next loading is not the same as an MPS histogram op limitation." + "reason": "Llava Next loading segfault and MPS histogram-not-implemented are separate platform/runtime issues." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "Different tokenizer vs image-classification bugs with no concrete overlap." + "reason": "Tokenizer crash on NER padding/truncation vs Perceiver image-classification failure at non-default resolution." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Tokenizer special-token registration issue vs a DAC latent/STE mismatch." + "reason": "BigBirdTokenizer special-token registration bug vs DAC.from_latents/forward mismatch; different subsystems." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Different subsystems: Qwen3-MoE sliding-window behavior vs AutoConfig trust_remote_code field loss." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to AutoConfig trust_remote_code missing fields." }, { - "left": "issue:43950", - "right": "issue:45987", + "left": "issue:43643", + "right": "issue:44877", "accept": false, - "reason": "Both involve regression-ish type issues, but one is buffer corruption and the other is a cache-length return-type bug." + "reason": "Both are config-loading issues, but one is missing fields from trust_remote_code and the other is strict config rejecting granite_speech; different causes." }, { - "left": "issue:43643", - "right": "issue:44877", + "left": "issue:43950", + "right": "issue:45987", "accept": false, - "reason": "Both config-related, but one is missing fields from trust_remote_code and the other is a strict config loading failure." + "reason": "Non-persistent buffer corruption in from_pretrained is unrelated to StaticCache.get_seq_length returning the wrong type." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Same helper name, but different root causes: offline model_info call vs backend_tokenizer attribute crash." + "reason": "Both touch Mistral regex patching, but one is an offline hub call in _patch_mistral_regex while the other is an AttributeError on backend_tokenizer; different failure points." }, { "left": "issue:30064", "right": "issue:36331", "accept": false, - "reason": "Completely unrelated image segmentation-processing and trainer API issues." + "reason": "Image segmentation-map processing and CustomTrainer.compute_loss signature mismatch are unrelated." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both are video-input bugs, but they affect different Qwen model lines and different failure modes." + "reason": "Same broad Qwen multimodal area, but different models and bugs: Qwen3OmniMoe video input handling vs qwen2.5-vl vision_position_ids." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Both are v5 behavior regressions, but on different models and code paths." + "reason": "Different models and regressions: Pegasus output changes across versions vs CamemBERT masked-LM prediction errors." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Different tokenizer/processor behaviors: post-processor update vs batch_decode skip_special_tokens." + "reason": "Both concern token/processor special-token behavior, but one is runtime post-processor updates and the other is Whisper batch_decode skipping special tokens." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Unrelated GPU/device-mapping crash vs tokenizer-loading regression." - }, - { - "left": "issue:43577", - "right": "issue:43818", - "accept": false, - "reason": "Different model families and different failures; only broad multimodal similarity." + "reason": "Different domains: Qwen3 auto device-mapping CUDA assert on A800 vs tokenizer loading failure for EMBEDDIA/est-roberta." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "Documentation/example issues are unrelated to an OCR index error." + "reason": "Documentation/example errors for SigLIP2 are not the same as Kosmos2.5's runtime index error on long OCR input." }, { - "left": "issue:43782", - "right": "issue:44977", + "left": "issue:43577", + "right": "issue:43818", "accept": false, - "reason": "Different Qwen model variants and unrelated generation/load failures." + "reason": "Different models and bug classes: Blip2 dtype mismatch vs Video-LLaVA missing temporal attention/shared weights." }, { - "left": "issue:44821", - "right": "issue:45084", + "left": "issue:43782", + "right": "issue:44977", "accept": false, - "reason": "AutoImageProcessor URL loading and template compilation are unrelated." + "reason": "Related model family, but one is a weight_only loading error and the other is a flash-attention generation bug; not mergeable as one fix." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Whisper empty-transcription generation and Kosmos2.5 OCR indexing are unrelated." - }, - { - "left": "issue:44186", - "right": "issue:44442", - "accept": false, - "reason": "Tokenizer crash on NER inputs is not the same as AutoTokenizer failing to load a specific tokenizer." + "reason": "Whisper empty-transcription generation after align_special_tokens is unrelated to Kosmos2.5 long-input indexing." }, { - "left": "issue:43644", - "right": "issue:43749", + "left": "issue:44821", + "right": "issue:45084", "accept": false, - "reason": "Both are loading regressions, but one corrupts buffers and the other breaks CPU RAM-efficient loading." + "reason": "AutoImageProcessor URL loading and TypeError compiling non-template nodes are unrelated code paths." }, { - "left": "issue:44561", - "right": "issue:45020", + "left": "issue:44186", + "right": "issue:44442", "accept": false, - "reason": "Both mention remote_code/trust_remote_code regressions, but 44561 is a specific removed-helper breakage rather than the broad umbrella in 45020." + "reason": "LayoutLMv2 tokenizer crash and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are different tokenizer/loading bugs." } ] }, @@ -5167,6 +5049,7 @@ "issue:43577", "issue:43582", "issue:43606", + "issue:43644", "issue:43650", "issue:43653", "issue:43697", @@ -5186,6 +5069,7 @@ "issue:44485", "issue:44509", "issue:44554", + "issue:44561", "issue:44610", "issue:44671", "issue:44805", @@ -5195,10 +5079,8 @@ "issue:45020", "issue:45072", "issue:45081", - "issue:45084", "issue:45325", "issue:45406", - "issue:45538", "issue:45701", "issue:45834", "issue:45987", @@ -5206,231 +5088,238 @@ "issue:46209" ], "soft_pairs": [ + "issue:43644|issue:43749", "issue:43606|issue:45072", "issue:44479|issue:45834", + "issue:44561|issue:45020", "issue:34634|issue:36010", - "issue:43825|issue:43994", "issue:43577|issue:43582", + "issue:43825|issue:43994", "issue:44805|issue:44936", - "issue:45987|issue:46192", "issue:44485|issue:44509", + "issue:45987|issue:46192", "issue:43825|issue:43827", "issue:44671|issue:45701", - "issue:44610|issue:45072", "issue:28218|issue:29159", + "issue:44610|issue:45072", "issue:43720|issue:45072", - "issue:43906|issue:46209", "issue:43782|issue:45325", + "issue:43906|issue:46209", "issue:42175|issue:45081", "issue:43749|issue:44291", - "issue:45081|issue:45406", "issue:43582|issue:44857", + "issue:45081|issue:45406", "issue:43262|issue:43819", - "issue:43526|issue:43697", "issue:34634|issue:36331", + "issue:43526|issue:43697", "issue:43650|issue:43906", "issue:44060|issue:44368", "issue:43653|issue:44912", "issue:43742|issue:45020", - "issue:43526|issue:44554", - "issue:45084|issue:45538" + "issue:43526|issue:44554" ], - "prepared_review_unit_hash": "612df665a5a0c28d15ad9f8ccdcb94a9abaff42d2e69aeb05aec1604721c47d6", + "prepared_review_unit_hash": "bcba21316dbae4e5fcea5ca8bd0ecd74e4889e4b5e35f707f4ff09bb9f8f6386", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36262, - "estimated_input_tokens": 9066, - "estimated_eval_tokens": 18388 + "soft_pair_count": 29, + "serialized_chars": 36609, + "estimated_input_tokens": 9153, + "estimated_eval_tokens": 18562 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated transformers bug reports that only share broad subsystem similarity (tokenizers, loading, multimodal processors, training, and v5 regressions). Under a strict duplicate standard, none of the soft-edge pairs should be merged.", + "summary": "This is a heterogeneous cluster with several distinct bug families (tokenizer behavior, pipeline/v5 docs, remote_code regressions, model-specific loading/inference bugs, and training/runtime errors). None of the soft pairs look like true duplicates; most are only thematically similar.", "confidence": 0.84, - "canonical_issue_reason": "No single canonical issue: the cluster spans distinct bugs across tokenizer behavior, model loading, processor logic, and training/runtime failures, so there isn\u2019t one underlying defect that the others clearly duplicate.", + "canonical_issue_reason": "No single canonical issue cleanly represents the whole cluster. If one must be chosen, issue 43825 is the most concrete representative of the small pipeline/v5 subset, but the cluster is too mixed for a strong canonical.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45701 is the broadest representative of the tokenizer-regression subset, but it still does not subsume the rest of the cluster and should not be treated as a duplicate hub.", + "best_issue_reason": "Issue 43825 is the clearest, most actionable bug report among the near-v5/pipeline items, but it still does not generalize to the unrelated tokenizer/model/training issues in the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Different failure modes: junk non-persistent buffers during save/load vs broken FSDP CPU RAM efficient loading. Same broad loading area, not the same bug." + }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "Both involve inference/runtime behavior, but one is a CPU-offload device mismatch and the other is a bfloat16 dtype mismatch; different failure modes and code paths." + "reason": "Both mention device/dtype issues during inference, but they affect different models and code paths (CPU offload vs bfloat16 inference). Not the same underlying defect." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Both are multimodal regressions, but one is a video-input bug in Qwen VL models and the other is a long-OCR input index error in Kosmos2.5; not the same defect." + "reason": "Video-input regression for Qwen VL models is unrelated to Kosmos2.5 OCR index errors. Different subsystems and symptoms." }, { - "left": "issue:34634", - "right": "issue:36010", + "left": "issue:44561", + "right": "issue:45020", "accept": false, - "reason": "BarkProcessor voice_preset failure and GenerationMixin import error are unrelated issues in different components." + "reason": "Both involve recent v5/remote_code breakage, but one is specifically `is_torch_fx_available` removal while the other is a broad regression report. Too general to merge as duplicates." }, { - "left": "issue:43825", - "right": "issue:43994", + "left": "issue:34634", + "right": "issue:36010", "accept": false, - "reason": "One is a pipeline error message regression about removed translation tasks; the other is a SigLIP2 model producing bad outputs with AutoModel/pipeline. Different bugs." + "reason": "BarkProcessor voice_preset failure and GenerationMixin import error are unrelated model/API issues." }, { "left": "issue:43577", "right": "issue:43582", "accept": false, - "reason": "Blip2 dtype propagation and Apple Silicon caching_allocator_warmup TypeError are unrelated." + "reason": "One is a BLIP2 dtype propagation bug; the other is an Apple Silicon allocator warmup TypeError. Same platform theme only." }, { - "left": "issue:44805", - "right": "issue:44936", + "left": "issue:43825", + "right": "issue:43994", "accept": false, - "reason": "Mask/index shape mismatch in one case versus trainer.evaluate() failing after train() in the other; no shared underlying bug." + "reason": "Pipeline translation-task messaging and SigLIP2 nonsensical outputs are unrelated problems." }, { - "left": "issue:45987", - "right": "issue:46192", + "left": "issue:44805", + "right": "issue:44936", "accept": false, - "reason": "One is a return type contract bug in StaticCache, the other is kwargs handling silently ignoring return_tensors; different APIs and failures." + "reason": "Generic mask/index shape crash versus trainer.evaluate() failing after train(). Different stack traces and code paths." }, { "left": "issue:44485", "right": "issue:44509", "accept": false, - "reason": "GLM-5 RoPE implementation discussion and docs mentioning removed pipeline tasks are unrelated." + "reason": "GLM-5 RoPE discussion is unrelated to docs still mentioning removed pipeline tasks." + }, + { + "left": "issue:45987", + "right": "issue:46192", + "accept": false, + "reason": "StaticCache sequence-length contract bug and ignored return_tensors in text_kwargs are separate API issues." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "Both touch v5 pipeline deprecations, but one is an incorrect error message and the other is stale docs; not the same code-path problem." + "reason": "They are related to the v5 pipeline removals, but one is an error-message bug and the other is stale docs. Same topic, not the same bug." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "CamemBERT masked-LM prediction regression and a general tokenization-change report are not the same concrete bug." - }, - { - "left": "issue:44610", - "right": "issue:45072", - "accept": false, - "reason": "Processor image-size mismatch versus dtype mismatch in inference; different model paths and symptoms." + "reason": "CamemBERT masked-LM regression and generic tokenization changes are not the same concrete defect." }, { "left": "issue:28218", "right": "issue:29159", "accept": false, - "reason": "Both are tokenizer-related, but one is about an extra space after an added token while the other is a broad slow/fast tokenizer inconsistency; not clearly the same bug." + "reason": "Both involve tokenizer behavior, but one is about an added token inserting extra space and the other is slow/fast tokenizer inconsistency more broadly. Not clearly the same bug." }, { - "left": "issue:43720", + "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "Packed-weight unpacking during accelerate loading and dtype mismatch in inference are unrelated loading/runtime issues." + "reason": "OmDet-Turbo processor size mismatch is unrelated to SwitchTransformers/TimmWrapperModel bfloat16 dtype mismatches." }, { - "left": "issue:43906", - "right": "issue:46209", + "left": "issue:43720", + "right": "issue:45072", "accept": false, - "reason": "An isolated reproduction of an earlier issue and an FP8 inference failure on B200 are unrelated." + "reason": "Packed-weight unpacking failure during accelerate loading is a different issue from dtype mismatches in inference." }, { "left": "issue:43782", "right": "issue:45325", "accept": false, - "reason": "Qwen3VL weight_only loading error and Qwen2.5-VL RoPE position-id scaling are separate model-specific bugs." + "reason": "Qwen3VL weight_only loading error and Qwen2.5-VL RoPE scaling bug are separate model-specific regressions." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "A reproduction of an older issue and an FP8/B200 inference bug are unrelated." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Backend dependency packaging issue versus tokenizer regex patch crash; no shared defect." + "reason": "Backend packaging issue for `.[torch]` and a Mistral regex patch crash are unrelated." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM efficient loading broken and init_empty_weights rejecting _is_hf_initialized are both loading-related but distinct failures." + "reason": "CPU RAM efficient loading breakage and unexpected `_is_hf_initialized` arg during init_empty_weights are distinct loading bugs." }, { - "left": "issue:45081", - "right": "issue:45406", + "left": "issue:43582", + "right": "issue:44857", "accept": false, - "reason": "Both are attribute errors during model/processor loading, but one is Mistral tokenizer regex patching and the other is Gemma4Processor missing _tokenizer; different code paths." + "reason": "AppleSilicon allocator warmup TypeError and CUDA float16 AMP crash in LwDetrImageLoss are different platform/runtime failures." }, { - "left": "issue:43582", - "right": "issue:44857", + "left": "issue:45081", + "right": "issue:45406", "accept": false, - "reason": "Apple Silicon warmup TypeError and CUDA AMP crash in LwDetrImageLoss are unrelated." + "reason": "Tokenizer regex patch crash and Gemma4Processor missing `_tokenizer` are separate loading/processor bugs." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Sampling-rate default in audio chat template and DAC.from_latents mismatch with missing STE are different audio bugs." + "reason": "apply_chat_template sampling-rate default bug and DAC.from_latents mismatch are unrelated audio issues." }, { - "left": "issue:43526", - "right": "issue:43697", + "left": "issue:34634", + "right": "issue:36331", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels bug and RTDetrV2 output differences in v5 are distinct vision regressions." + "reason": "BarkProcessor voice_preset does not work and CustomTrainer.compute_loss signature breakage are unrelated." }, { - "left": "issue:34634", - "right": "issue:36331", + "left": "issue:43526", + "right": "issue:43697", "accept": false, - "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss unexpected keyword argument are unrelated." + "reason": "BeitImageProcessorFast label reduction bug and RTDetrV2 output regression are different model/processor defects." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "A vague request title and an isolated reproduction of another issue do not indicate the same bug." + "reason": "'ADD THE DATA' is not a technical bug report, and it is not the same as the referenced reproduction issue." }, { "left": "issue:44060", "right": "issue:44368", "accept": false, - "reason": "Incorrect tied-weights warning in Qwen3-Next and a separate warning about tie_word_embeddings in fine-tuning are related only by theme, not the same defect." + "reason": "Incorrect tied-weights warning in Qwen3-Next and a Qwen3.5 LoRA warning are related only by theme, not by identical code path." }, { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "BigBirdTokenizer special-token decode bug and MXFP4 quantization fallback are unrelated." + "reason": "BigBirdTokenizer special-token registration bug and MXFP4 quantization fallback are unrelated." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "KeyError loading MobileLLM-125M and remote_code breakage across recent versions are different loading problems." + "reason": "KeyError loading MobileLLM-125M and broad remote_code breakage are different failures." }, { "left": "issue:43526", "right": "issue:44554", "accept": false, - "reason": "BeitImageProcessorFast label handling and an MPS attention correctness issue are unrelated." - }, - { - "left": "issue:45084", - "right": "issue:45538", - "accept": false, - "reason": "Template compilation error and CLIPTokenizer model_max_length behavior are different tokenizer/framework issues." + "reason": "BeitImageProcessorFast label reduction and MPS attention correctness with differing head dims are unrelated." } ] }, @@ -5454,7 +5343,6 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ - "issue:29159", "issue:34567", "issue:36010", "issue:36331", @@ -5462,7 +5350,6 @@ "issue:39692", "issue:40990", "issue:42490", - "issue:42915", "issue:43257", "issue:43262", "issue:43329", @@ -5478,17 +5365,17 @@ "issue:43645", "issue:43696", "issue:43720", - "issue:43742", "issue:43746", "issue:43749", "issue:43819", "issue:43854", "issue:43927", - "issue:43931", + "issue:44387", "issue:44479", "issue:44568", "issue:44589", "issue:44610", + "issue:44617", "issue:44756", "issue:44811", "issue:44912", @@ -5496,28 +5383,32 @@ "issue:44977", "issue:44991", "issue:45042", + "issue:45084", "issue:45356", "issue:45459", + "issue:45538", "issue:45600", "issue:45759", "issue:45803", "issue:45859" ], "soft_pairs": [ + "issue:45084|issue:45538", "issue:43329|issue:45600", "issue:44589|issue:45042", "issue:43746|issue:44811", "issue:43638|issue:44936", "issue:43452|issue:44991", + "issue:44387|issue:44617", "issue:43550|issue:44912", "issue:36010|issue:43645", "issue:43819|issue:43927", "issue:43257|issue:43329", - "issue:45459|issue:45803", "issue:39692|issue:44912", "issue:43854|issue:44977", "issue:38175|issue:44610", "issue:43720|issue:44610", + "issue:45459|issue:45803", "issue:36010|issue:36331", "issue:43454|issue:43577", "issue:43531|issue:45042", @@ -5527,222 +5418,213 @@ "issue:42490|issue:43257", "issue:43643|issue:45759", "issue:43329|issue:43450", - "issue:43749|issue:44756", "issue:43606|issue:45859", + "issue:43749|issue:44756", "issue:34567|issue:40990", - "issue:44568|issue:45356", - "issue:43257|issue:43931", - "issue:29159|issue:34567", - "issue:42915|issue:43742" + "issue:44568|issue:45356" ], - "prepared_review_unit_hash": "8096d6657a3045855883c0ba3a840ceacff24cd745fd37df9e479b399e9b6766", + "prepared_review_unit_hash": "66261d035057627a14bec6a78bf5ef57a6a2352f24533ae58a55fea168cd28fa", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36938, - "estimated_input_tokens": 9235, - "estimated_eval_tokens": 18726 + "soft_pair_count": 29, + "serialized_chars": 36621, + "estimated_input_tokens": 9156, + "estimated_eval_tokens": 18568 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: tokenizer, trainer, multimodal/video, quantization, loading, and CPU-offload issues are mixed together. The soft-similarity links look topical rather than truly duplicate, so none should be merged as the same underlying bug/change.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43257 is the most connected item in the soft-pair graph and the clearest representative of the Qwen/MoE loading subset, with a concrete reproducible failure description.", + "summary": "The cluster is highly heterogeneous: the soft pairs are only superficially related by subsystem/theme, not by the same underlying bug or fix. No PRs are present.", + "confidence": 0.94, + "canonical_issue_reason": "No true canonical issue exists here; the items span many unrelated bugs (training, tokenization, multimodal/video, loading, offload, quantization, OOM), so any single issue would be a poor duplicate representative.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43257 is the best anchor for this cluster because it is central to several nearby Qwen/MoE-related reports, while still being a specific, well-scoped bug.", + "best_issue_reason": "issue:43749 is the strongest single anchor only because it has the most discussion/inbound references, but it still does not represent the rest of this mixed cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated bugs: template-node compilation vs tokenizer max-length constant." + }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Different bugs: multimodal video token counting uses undefined helpers, while the other is a removed-config reference in auto_mappings/CI." + "reason": "Different code paths: multimodal video token counting vs stale auto_mappings references." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Unrelated failure modes: float8 storage lookup during loading vs PIL image processors incorrectly requiring torchvision." + "reason": "Different failures: missing Float8 storage vs PIL image processor torchvision requirement." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "Different code paths: local PEFT checkpoint loading vs Whisper processor batch_decode ignoring skip_special_tokens." + "reason": "Different features and code paths: local PEFT checkpoint loading vs Whisper batch_decode behavior." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Different trainer failures: Deepspeed ZeRO-3 indexing with a non-pretrained Bert model vs evaluate() breaking after train()." + "reason": "Both are training/eval failures, but the underlying bugs are different (DeepSpeed ZeRO-3 index error vs evaluate-after-train failure)." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "Both are loading issues, but the concrete defects differ: gguf_file handling vs tokenizer loading regression for est-roberta." + "reason": "Different model loading problems: gguf_file handling vs tokenizer loading for a specific repo." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both are OOM reports, but for different models and mechanisms; not the same bug." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Different model/runtime problems: torch.compile SDPA crash on Bamba vs MXFP4 quantization fallback on gpt-oss-20b." + "reason": "Unrelated: torch.compile/SDPA crash in Bamba vs MXFP4 load fallback in gpt-oss." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "Both involve import/init behavior, but one is a GenerationMixin import error and the other is a Jupyter custom-model initialization regression." + "reason": "Different regressions: GenerationMixin import path vs custom model initialization in Jupyter." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "Different model classes and bugs: DAC from_latents/STE mismatch vs DiaConfig losing custom token IDs on save/load." + "reason": "Different model bugs: DAC latent reconstruction mismatch vs DiaConfig token-id persistence." }, { "left": "issue:43257", "right": "issue:43329", "accept": false, - "reason": "Related ecosystem area, but not the same defect: Qwen3 MoE weight conversion/loading vs undefined video token helper usage." - }, - { - "left": "issue:45459", - "right": "issue:45803", - "accept": false, - "reason": "Both touch error handling, but one hides tokenizer decode errors and the other uses bare except in Fuyu tensor conversion." + "reason": "Different areas: Qwen3 MoE loading under accelerate/deepspeed vs multimodal video token counting." }, { "left": "issue:39692", "right": "issue:44912", "accept": false, - "reason": "Different scopes: documentation example errors for SigLIP2 vs a quantization/loading issue on gpt-oss-20b." + "reason": "Different issues: SigLIP2 docs/example problems vs gpt-oss MXFP4 loading behavior." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "Different problems: GLM-4.7-Flash loading in tests vs Qwen3.5 generation failure with flash-attention." + "reason": "Different model-specific generation/load failures; not the same concrete bug." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Different model behaviors: unexpected zero probabilities in SigLIP2 vs processor/model input-size mismatch in OmDet-Turbo." + "reason": "Different vision-model issues: zero probabilities vs processor/model input-size mismatch." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "Different defects: packed weights not unpacked during accelerate loading vs wrong processor output shape for batched video inputs." + "reason": "Different code paths: packed-weight loading vs processor output shape." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve exception handling, but in different functions and with different bugs." }, { "left": "issue:36010", "right": "issue:36331", "accept": false, - "reason": "Completely different trainer/import failures: GenerationMixin import error vs custom compute_loss signature mismatch." + "reason": "Both affect trainer-related APIs, but one is an import error and the other is a compute_loss signature break." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "Different model bugs: AyaVision lm_head weight tying vs BLIP2 dtype not propagating to submodules." + "reason": "Different model loading bugs: lm_head weight tying vs dtype propagation to QFormer." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Different subsystems: Qwen3-MoE sliding_window behavior vs PIL backend image processors requiring torchvision." + "reason": "Unrelated: Qwen3-MoE sliding-window logic vs image processor torchvision dependency." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Both are processor issues, but one is audio chat-template sample-rate defaulting and the other is Whisper decode option handling." + "reason": "Both are audio/processor issues, but sampling-rate defaulting and batch_decode special-token handling are distinct." }, { "left": "issue:43257", "right": "issue:43696", "accept": false, - "reason": "Different symptoms: Qwen3 MoE loading/weight conversion vs plain CUDA OOM on gpt-oss-20b." + "reason": "Different problems: MoE weight conversion under accelerate/deepspeed vs plain OOM on GPT-oss-20b." }, { "left": "issue:43329", "right": "issue:44479", "accept": false, - "reason": "Both concern video VLMs, but one is an undefined helper in token counting and the other is a broader v5.3.0 video-input regression." + "reason": "Both are video-related, but one is undefined variables in token counting and the other is a broader v5.3.0 regression." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Unrelated loading issues: slash-suffixed local path warning vs Qwen3 MoE weight conversion with accelerate+deepspeed." + "reason": "Completely different topics: trailing-slash local-path warning vs Qwen3 MoE loading." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "Both are config-related, but one loses fields with trust_remote_code and the other fails to unwrap text_config for composite multimodal configs." + "reason": "Both are config-loading issues, but one is missing fields with trust_remote_code and the other is text_config unwrapping for multimodal configs." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Same broad video area, but different bugs: undefined video token helpers vs incorrect batched video processor output shape." + "reason": "Different video bugs: undefined helper variables vs incorrect batched processor output shape." }, { - "left": "issue:43749", - "right": "issue:44756", + "left": "issue:43606", + "right": "issue:45859", "accept": false, - "reason": "Both mention loading/memory, but FSDP RAM-efficient loading is a different failure from disabling mmap to avoid Strix Halo OOMs." + "reason": "Both mention CPU offload, but they are different models and failure modes." }, { - "left": "issue:43606", - "right": "issue:45859", + "left": "issue:43749", + "right": "issue:44756", "accept": false, - "reason": "Both involve CPU offload, but they affect different models and different code paths; no evidence of the same underlying bug." + "reason": "Different loading/memory issues: FSDP CPU RAM efficient loading vs mmap on Strix Halo OOM." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "Different training metrics/perplexity issues with no shared failure mechanism." + "reason": "Unrelated: TrainerState token counter not updating vs GPT-oss perplexity issue." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Both tokenizer regressions, but one is add_special_tokens/BOS-EOS behavior and the other is codec handling plus warning text." - }, - { - "left": "issue:43257", - "right": "issue:43931", - "accept": false, - "reason": "Different model families and failures: Qwen3 MoE loading/weight conversion vs Qwen3-VL shape mismatch on load." - }, - { - "left": "issue:29159", - "right": "issue:34567", - "accept": false, - "reason": "Unrelated issues: tokenizer slow/fast inconsistency vs TrainerState token-count tracking not updating." - }, - { - "left": "issue:42915", - "right": "issue:43742", - "accept": false, - "reason": "Different loading/config bugs: FineGrainedFP8Config failure on Qwen3Moe vs key error loading MobileLLM-125M." + "reason": "Both are tokenizer regressions, but for different models and different root causes." } ] }, @@ -5779,7 +5661,6 @@ "issue:42915", "issue:43257", "issue:43317", - "issue:43404", "issue:43452", "issue:43540", "issue:43577", @@ -5791,6 +5672,7 @@ "issue:43782", "issue:43824", "issue:43927", + "issue:43931", "issue:44062", "issue:44291", "issue:44479", @@ -5812,17 +5694,20 @@ "issue:45488", "issue:45538", "issue:45701", - "issue:45907", - "issue:45987" + "issue:45823", + "issue:45907" ], "soft_pairs": [ + "issue:43257|issue:43931", + "issue:29159|issue:34567", + "issue:42915|issue:43742", "issue:44291|issue:45042", "issue:42915|issue:44821", "issue:44849|issue:45701", "issue:44568|issue:45488", "issue:44987|issue:45538", - "issue:43697|issue:45357", "issue:43452|issue:43582", + "issue:43697|issue:45357", "issue:41720|issue:43540", "issue:43927|issue:44912", "issue:42915|issue:45538", @@ -5833,234 +5718,252 @@ "issue:43540|issue:44560", "issue:43317|issue:45538", "issue:44062|issue:44991", - "issue:44987|issue:45907", "issue:45137|issue:45405", + "issue:44987|issue:45907", "issue:24643|issue:29159", - "issue:43257|issue:43756", "issue:28282|issue:34689", "issue:43577|issue:45072", + "issue:43257|issue:43756", "issue:43742|issue:44291", - "issue:42175|issue:44488", "issue:28282|issue:29127", + "issue:42175|issue:44488", "issue:42915|issue:43749", "issue:38175|issue:44898", "issue:43582|issue:43782", - "issue:43404|issue:45987" + "issue:43540|issue:45823" ], - "prepared_review_unit_hash": "f143388261b7a4345e61aba68e543d55b8ade03dc655d14bb74bc1027e3be54d", + "prepared_review_unit_hash": "ce942500b578f62cb2daa85ac56ab686d0914f14c747e94ce25803ec9bfad0b2", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36723, - "estimated_input_tokens": 9181, - "estimated_eval_tokens": 18618 + "soft_pair_count": 33, + "serialized_chars": 37558, + "estimated_input_tokens": 9390, + "estimated_eval_tokens": 19036 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated Transformers bug reports spanning tokenizer, vision/video, loading, quantization, and training regressions. The only plausible duplicate pair is the Qwen2.5-VL video-input regression reports; the rest are distinct issues.", - "confidence": 0.74, - "canonical_issue_reason": "issue:44479 is the strongest canonical anchor: it describes a concrete v5.3.0 video-input regression affecting multiple Qwen VL variants, and it matches the only clear duplicate-like pair in this set.", + "summary": "This cluster is mostly heterogeneous: tokenizer regressions, model-loading errors, video-input bugs, dtype/device-map issues, and a few DeepSpeed/quantization reports. I only see one plausible duplicate pair in the Qwen VL video-input regression family.", + "confidence": 0.61, + "canonical_issue_reason": "Issue 45701 is the broadest and most representative anchor here: it explicitly frames a cross-version tokenization change, which is the closest shared theme among several tokenizer-related reports in an otherwise mixed set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the most actionable and broadly scoped real regression report in the cluster; it captures the shared video-input bug better than the narrower follow-up report.", + "best_issue_reason": "45701 is the best single issue to represent the cluster because it is generic, active, and closest to the recurring regression theme; the other items are mostly narrower or unrelated one-off failures.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Both are Qwen loading problems, but one is an MoE weight-conversion issue under accelerate+deepspeed and the other is a Qwen3-VL weight-shape mismatch; different failure modes and code paths." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer inconsistency and TrainerState token-count tracking are unrelated subsystems." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3Moe FP8 loading failure vs MobileLLM key error are different model-loading bugs." + }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "Different failures in different subsystems: init_empty_weights/_is_hf_initialized vs PIL image processors requiring torchvision." + "reason": "Init-empty-weights argument handling and PIL/torchvision backend requirements are unrelated." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Unrelated: Qwen3Moe FineGrainedFP8Config error versus AutoImageProcessor URL loading." + "reason": "Qwen3Moe FP8 loading is not the same bug as AutoImageProcessor URL loading." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Different bugs: Qwen3.5 hidden-state handling vs a broad tokenization-version regression." + "reason": "Qwen3.5 hidden-state output bug is unrelated to the tokenization-version regression in 45701." }, { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both involve tokenizers, but they are different concrete bugs on different models/code paths and not one fix." + "reason": "Both concern tokenizer behavior, but they affect different tokenizers and different mechanisms (BOS/EOS insertion vs LlamaTokenizer pre-tokenizer override)." }, { "left": "issue:44987", "right": "issue:45538", "accept": false, - "reason": "Different issue classes: loading physical-intelligence/fast vs CLIPTokenizer model_max_length." + "reason": "Loading failure for a specific model and CLIPTokenizer max-length behavior are not the same underlying bug." }, { - "left": "issue:43697", - "right": "issue:45357", + "left": "issue:43452", + "right": "issue:43582", "accept": false, - "reason": "Separate model-specific bugs: RTDetrV2 output drift vs Qwen3.5 save_pretrained key mapping." + "reason": "gguf_file loading breakage and Apple Silicon allocator warmup TypeError are unrelated." }, { - "left": "issue:43452", - "right": "issue:43582", + "left": "issue:43697", + "right": "issue:45357", "accept": false, - "reason": "Completely unrelated: gguf tokenizer/model loading vs Apple Silicon allocator warmup TypeError." + "reason": "RTDetr output divergence and Qwen3.5 save_pretrained key corruption are different code paths." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Same family name, but different failures and code paths: auto device mapping CUDA assert vs video-input ValueError." + "reason": "Auto device-mapping CUDA assert on Qwen3 is unrelated to Qwen3OmniMoe video processing." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "Different areas entirely: DiaConfig save/load token IDs vs MXFP4 quantization fallback." + "reason": "Custom token ID persistence and MXFP4 quantization fallback are unrelated." }, { "left": "issue:42915", - "right": "issue:45538", + "right": "issue:43749", "accept": false, - "reason": "Unrelated: Qwen3Moe FP8 config failure versus CLIP tokenizer max-length behavior." + "reason": "FP8 Qwen3Moe loading and FSDP CPU RAM efficient loading are different bugs." }, { "left": "issue:44479", "right": "issue:45381", "accept": true, - "reason": "Both report the same v5.3.0 Qwen2.5-VL video-input regression around incorrect vision position IDs, so they likely share one fix." + "reason": "Both report the same Qwen VL video-input regression in v5.3.0, centered on incorrect video-position handling; they look like the same underlying code-path issue." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "Different bugs: Mac bus error on a CLIP community model vs TrainerState token counting not updating." + "reason": "MacOS CLIP bus error and TrainerState token counting are unrelated." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both concern video inputs, but the model families and symptoms differ enough that this is not clearly one underlying bug." + "reason": "Qwen3OmniMoe video ValueError and Qwen2.5/3.x VL video regression are different model families and likely different bugs." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "Unrelated packaging/import problems: missing TensorFlow backend vs missing Qwen2_5_VL class export." + "reason": "Packaging/backend dependency inclusion and missing Qwen2_5_VL import are unrelated." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Different Qwen video bugs with different models and error modes; not the same concrete fix." + "reason": "Both are Qwen video issues, but they describe different models and different error types." }, { "left": "issue:43317", "right": "issue:45538", "accept": false, - "reason": "Unrelated: device_map/offload dequantized loading vs CLIPTokenizer max-length constant." + "reason": "Device-map/offload dequantization failure and CLIP tokenizer max-length behavior are unrelated." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "Tokenizer-related, but the failure modes are different and model-specific; too weak to treat as the same bug." - }, - { - "left": "issue:44987", - "right": "issue:45907", - "accept": false, - "reason": "Different concerns: model loading regression for physical-intelligence/fast vs list_repo_files revision forwarding." + "reason": "AddedToken special-arg duplication and tokenizer loading for est-roberta are different tokenizer bugs." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "Completely unrelated: DeepSpeed ZeRO3 deque underflow vs a PEFT version pin issue." + "reason": "DeepSpeed ZeRO3 deque error and PEFT version bump are unrelated." }, { - "left": "issue:24643", - "right": "issue:29159", + "left": "issue:44987", + "right": "issue:45907", "accept": false, - "reason": "Different problems: DeepSpeed training weight shape error vs slow/fast tokenizer inconsistency." + "reason": "Model loading regression and list_repo_files revision forwarding are unrelated." }, { - "left": "issue:43257", - "right": "issue:43756", + "left": "issue:24643", + "right": "issue:29159", "accept": false, - "reason": "Both are model bugs, but one is Qwen3 MOE weight conversion and the other is Smollm3 RoPE-layer behavior." + "reason": "DeepSpeed training weight-shape error and tokenizer slow/fast inconsistency are unrelated." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "Import/runtime availability issue versus a specific model-loading regression; not the same underlying bug." + "reason": "Missing PyTorch import and Llama 3.2 Vision loading regression are different issues." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Both mention dtype, but they affect different models and different inference paths." + "reason": "Both mention dtype, but one is BLIP2 loading state and the other is CI inference mismatches in different models." }, { - "left": "issue:43742", - "right": "issue:44291", + "left": "issue:43257", + "right": "issue:43756", "accept": false, - "reason": "Different loading failures: key error for MobileLLM-125M vs unexpected init_empty_weights argument." + "reason": "Qwen3 MOE weight conversion and Smollm3 RoPE-layer count are unrelated." }, { - "left": "issue:42175", - "right": "issue:44488", + "left": "issue:43742", + "right": "issue:44291", "accept": false, - "reason": "Different issues: missing TensorFlow backend in one install path vs a specific model failing to load." + "reason": "MobileLLM key error and init_empty_weights constructor mismatch are different failures." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Missing PyTorch import error and LayoutLMv3 box-validation messaging are unrelated." + "reason": "AutoModel missing PyTorch and LayoutLMv3 error-message clarity are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "TensorFlow optional dependency packaging and a specific model-loading failure are unrelated." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Different failures in different subsystems: Qwen3Moe FP8 config vs FSDP CPU RAM-efficient loading." + "reason": "Same as the earlier Qwen3Moe FP8 vs FSDP mismatch: different bugs." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "Different image-model bugs with different failure modes; no clear shared code-path." + "reason": "Unexpected zero probabilities in SigLIP2 and Perceiver image-classification resolution handling are unrelated." }, { "left": "issue:43582", "right": "issue:43782", "accept": false, - "reason": "Unrelated: Apple Silicon allocator TypeError vs Qwen3VL weight_only loading error." + "reason": "Apple Silicon allocator TypeError and Qwen3VL weight_only loading error are unrelated." }, { - "left": "issue:43404", - "right": "issue:45987", + "left": "issue:43540", + "right": "issue:45823", "accept": false, - "reason": "Different bugs: Mistral3 lm_head tying issue vs StaticCache returning the wrong type." + "reason": "Qwen3OmniMoe video processing and Gemma4 PLE device mismatch are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded overall. The summary stays conservative, and the only accepted soft-edge pair (44479 \u2194 45381) is well supported by the packet titles as the same Qwen2.5/Qwen3.x video-input regression family. The rest of the pairwise rejections are plausibly cautious and do not overclaim duplicate status." + "feedback": "The summary is grounded in the packet and stays conservative about the mostly heterogeneous cluster. The lone accepted pair (44479/45381) is plausibly the same Qwen VL video-input regression family, and the other soft-edge rejects do not overclaim duplication." }, "evaluator_used": true, "retried": false, @@ -6086,11 +5989,11 @@ "issue:34567", "issue:36331", "issue:41762", + "issue:43404", "issue:43452", "issue:43493", "issue:43526", "issue:43531", - "issue:43540", "issue:43643", "issue:43688", "issue:43697", @@ -6125,14 +6028,14 @@ "issue:45479", "issue:45588", "issue:45600", - "issue:45823", "issue:45920", + "issue:45987", "issue:46121", "issue:46202" ], "soft_pairs": [ "issue:44737|issue:45588", - "issue:43540|issue:45823", + "issue:43404|issue:45987", "issue:43526|issue:45479", "issue:43906|issue:45381", "issue:43493|issue:43994", @@ -6142,206 +6045,206 @@ "issue:44913|issue:46121", "issue:44062|issue:45081", "issue:44792|issue:44964", - "issue:45357|issue:45920", "issue:45042|issue:45081", - "issue:43452|issue:44488", + "issue:45357|issue:45920", "issue:43742|issue:44589", - "issue:43643|issue:45375", + "issue:43452|issue:44488", "issue:43531|issue:44821", + "issue:43643|issue:45375", "issue:45405|issue:45600", "issue:44625|issue:44877", "issue:24643|issue:29127", - "issue:43697|issue:45216", "issue:43720|issue:44898", + "issue:43697|issue:45216", "issue:43906|issue:46202", "issue:41762|issue:44936", - "issue:44514|issue:45381", - "issue:43688|issue:43749" + "issue:43688|issue:43749", + "issue:44514|issue:45381" ], - "prepared_review_unit_hash": "be26b2e5d2180622581963fed023c1fd1802c7d1b090256be2c7e9bc5b47171a", + "prepared_review_unit_hash": "b54cd1fabff219749c1fd7104c4fb35a6d09ed42fcba9cafd1123ba56bf47140", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35860, - "estimated_input_tokens": 8965, - "estimated_eval_tokens": 18186 + "serialized_chars": 35911, + "estimated_input_tokens": 8978, + "estimated_eval_tokens": 18212 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a noisy mix of unrelated Transformers bugs and regressions; none of the soft pairs look like the same concrete underlying issue, so I would not merge any of them as duplicates.", - "confidence": 0.56, - "canonical_issue_reason": "issue:45920 is the broadest and most representative issue in the set: it describes a v5 regression with multiple named affected models and has the strongest cross-link signal, making it the best anchor for a noisy cluster.", + "summary": "This set is heterogeneous and the soft pairs mostly share only broad subsystem names, not the same underlying defect. I would not merge any of the proposed soft edges. The most representative issue in the cluster is the broad tokenizer regression in 45920.", + "confidence": 0.24, + "canonical_issue_reason": "issue:45920 is the broadest, most cross-cutting report here: it describes a tokenizer regression affecting several model families, making it a better cluster representative than the many narrow, model-specific bugs.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45920 is the best issue to keep because it is clearly scoped, regression-oriented, and broad enough to represent the cluster\u2019s most general recurring theme.", + "best_issue_reason": "issue:45920 is the best global representative because it spans multiple affected models and points to a shared code-path regression rather than a one-off model-specific symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44737", "right": "issue:45588", "accept": false, - "reason": "Different bugs and code paths: XLNet CPU tensor placement vs flash-attention auxiliary-loss handling." + "reason": "XLNet device-placement work vs a flash-attention auxiliary-loss crash; unrelated code paths." }, { - "left": "issue:43540", - "right": "issue:45823", + "left": "issue:43404", + "right": "issue:45987", "accept": false, - "reason": "Different model families and failures: Qwen3OmniMoe video processing vs Gemma4 PLE device-map mismatch." + "reason": "Mistral3 weight tying vs StaticCache sequence-length return type; different bugs." }, { "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "One is BEiT label reduction logic; the other is sequence-classification loss degeneracy with num_labels=1." + "reason": "BeitImageProcessorFast label reduction vs sequence-classification loss degeneracy; not the same defect." }, { "left": "issue:43906", "right": "issue:45381", "accept": false, - "reason": "The former is only a reproduction of another issue; the latter is a specific Qwen2.5-VL video position-id bug, not the same failure." + "reason": "An isolated reproduction of another issue vs a Qwen2.5-VL video position-id bug; too different to treat as duplicates." }, { "left": "issue:43493", "right": "issue:43994", "accept": false, - "reason": "Both mention SigLIP2, but one is a HF-vs-JAX implementation discrepancy while the other is an AutoModel/pipeline output problem." + "reason": "Both concern SigLIP2, but one is HF-vs-JAX parity and the other is nonsensical pipeline output; not clearly the same code-path bug." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token accounting and a custom compute_loss signature error are unrelated training issues." + "reason": "TrainerState token counting vs a custom compute_loss signature error; unrelated trainer issues." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Different RoPE-related defects in different models: missing RoPE layers vs incorrect temporal position scaling." + "reason": "Smollm3 RoPE layer dropping vs Qwen2.5-VL temporal position scaling; distinct model-specific problems." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "Unrelated loading failures: init_empty_weights passes an unexpected arg, while AutoImageProcessor fails to load from URL." + "reason": "init_empty_weights/_is_hf_initialized mismatch vs AutoImageProcessor URL loading; different loading paths." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "Both touch rope/config handling, but one is config round-trip defaulting and the other is a TypeError in rope validation for JSON-loaded lists." + "reason": "GPTNeoXConfig reload resetting rotary_pct vs rope-validation TypeError on JSON-loaded lists; different serialization/validation bugs." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "Tokenizer crashes, but with different root causes: AddedToken kwarg duplication vs Mistral regex patch accessing missing backend_tokenizer." + "reason": "AddedToken kwarg duplication vs Mistral regex backend_tokenizer AttributeError; unrelated tokenizer failures." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Different multimodal model-load/test failures with no evidence of the same code-path or fix." - }, - { - "left": "issue:45357", - "right": "issue:45920", - "accept": false, - "reason": "One is a save_pretrained visual-encoder-key regression; the other is wrong token IDs from AutoTokenizer across several models." + "reason": "janus generation test failure vs Phi-4 multimodal model loading failure; different models and symptoms." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image-processor dependency behavior vs tokenizer regex patching; not the same bug." + "reason": "PIL backend torchvision dependency issue vs Mistral tokenizer regex patch crash; no shared bug." }, { - "left": "issue:43452", - "right": "issue:44488", + "left": "issue:45357", + "right": "issue:45920", "accept": false, - "reason": "Both involve loading models/tokenizers, but the reported failures are different and not clearly one underlying defect." + "reason": "Qwen3.5 save_pretrained visual-key regression vs tokenizer wrong token IDs across several models; different subsystems." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "A key error when loading a specific model is not the same as a missing Float8 storage type error." + "reason": "MobileLLM key error vs missing Float8 storage lookup; distinct loading failures." }, { - "left": "issue:43643", - "right": "issue:45375", + "left": "issue:43452", + "right": "issue:44488", "accept": false, - "reason": "One is a generic trust_remote_code config-field loss, the other is a specific Qwen config field being dropped by strict validation." + "reason": "gguf_file/from_pretrained handling vs failure to load cjvt/sleng-bert; not the same code path." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to AutoImageProcessor URL loading." + "reason": "Qwen3-MoE sliding_window behavior vs AutoImageProcessor URL loading; unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Generic trust_remote_code config-field loss vs Qwen3_5MoeVisionConfig strict field dropping; different config issues." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "A dependency version bump and stale auto_mappings references are different problems." + "reason": "Dependency-version bump vs stale auto_mappings references; unrelated maintenance problems." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Both are config-related, but num_labels propagation and strict config blocking granite_speech loading are distinct issues." - }, - { - "left": "issue:24643", - "right": "issue:29127", - "accept": false, - "reason": "DeepSpeed training weight-shape error and LayoutLMv3 box-info error are unrelated." + "reason": "num_labels propagation bug vs granite_speech config loading strictness; different config concerns." }, { - "left": "issue:43697", - "right": "issue:45216", + "left": "issue:43688", + "right": "issue:43749", "accept": false, - "reason": "RTDetrV2 output mismatch and Qwen3.5 save_pretrained regression are separate model-specific regressions." + "reason": "Auxiliary-loss normalization vs FSDP CPU RAM efficient loading; unrelated model/training bugs." }, { - "left": "issue:43720", - "right": "issue:44898", + "left": "issue:44514", + "right": "issue:45381", "accept": false, - "reason": "BitNet packed-weight loading and Perceiver interpolation are different model/path failures." + "reason": "Batched chat-template crash vs video vision_position_ids bug; distinct Qwen2.5-VL paths." }, { "left": "issue:43906", "right": "issue:46202", "accept": false, - "reason": "An isolated reproduction of another issue and a Gemma pipeline roundtrip failure are not the same concrete bug." + "reason": "Reproduction of an external issue vs Gemma pipeline roundtrip failure on a token boundary; not the same underlying bug." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "Gemma3 ZeRO-3 loading IndexError and trainer.evaluate failing after train are unrelated." + "reason": "Gemma3 ZeRO-3 loading error vs trainer.evaluate() failing after train(); different lifecycle stages and failures." }, { - "left": "issue:44514", - "right": "issue:45381", + "left": "issue:43720", + "right": "issue:44898", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is chat-template batching with padding=False and the other is video position-id computation; different code paths." + "reason": "BitNet packed-weight loading bug vs Perceiver interpolation failure; unrelated." }, { - "left": "issue:43688", - "right": "issue:43749", + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output regression vs Qwen3.5 save_pretrained regression; different models and code paths." + }, + { + "left": "issue:43906", + "right": "issue:45381", "accept": false, - "reason": "Aux-loss normalization and FSDP CPU RAM efficient loading are unrelated." + "reason": "Reproduction issue vs Qwen2.5-VL video position-id scaling bug; not the same defect." } ] }, @@ -6376,14 +6279,17 @@ "issue:41720", "issue:43295", "issue:43452", + "issue:43531", "issue:43577", "issue:43646", "issue:43653", "issue:43688", "issue:43697", + "issue:43742", "issue:43749", "issue:43819", "issue:43906", + "issue:43927", "issue:44361", "issue:44442", "issue:44464", @@ -6403,37 +6309,34 @@ "issue:45356", "issue:45372", "issue:45440", - "issue:45491", "issue:45542", "issue:45701", "issue:45704", "issue:45823", - "issue:46002", "issue:46050", "issue:46170", "issue:46178" ], "soft_pairs": [ "issue:45372|issue:46170", - "issue:43452|issue:43577", "issue:44466|issue:45005", - "issue:45542|issue:46178", + "issue:43452|issue:43577", "issue:44493|issue:46050", - "issue:43452|issue:44843", "issue:43819|issue:44811", + "issue:43452|issue:44843", "issue:30064|issue:33666", "issue:44936|issue:45823", + "issue:45542|issue:46178", "issue:44484|issue:44704", - "issue:43295|issue:45356", "issue:28282|issue:32090", + "issue:43295|issue:45356", "issue:43653|issue:44361", - "issue:44514|issue:45290", "issue:44361|issue:44442", "issue:28282|issue:33357", - "issue:44464|issue:44743", + "issue:44514|issue:45290", "issue:38175|issue:43295", + "issue:44464|issue:44743", "issue:28282|issue:35141", - "issue:45491|issue:46002", "issue:41720|issue:43906", "issue:43653|issue:44568", "issue:43577|issue:43697", @@ -6442,204 +6345,212 @@ "issue:43646|issue:43749", "issue:43688|issue:45440", "issue:29159|issue:33357", - "issue:43688|issue:45701" + "issue:43531|issue:43742", + "issue:43688|issue:45701", + "issue:43653|issue:43927" ], - "prepared_review_unit_hash": "3989d3d82a07f87833ec5da8b6e01c1616f27ad19fac16860be80df9a68640fa", + "prepared_review_unit_hash": "ab4bd63918f77fb5ecb320405a4a7250cdcfbb2de872fda6065cbda30c5cb6a1", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 36025, - "estimated_input_tokens": 9007, - "estimated_eval_tokens": 18270 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36841, + "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 set of unrelated Transformers bugs that only share broad surface keywords like tokenizer, loading, tied weights, or generation. None of the soft pairs looks like a definite duplicate; they appear to be separate regressions or edge cases in different code paths.", - "confidence": 0.93, - "canonical_issue_reason": "Issue #43452 is the best representative issue here: it is a concrete, reproducible loading regression with a clear trigger (`gguf_file`) and broad impact across both AutoTokenizer and AutoModelForCausalLM instantiation.", + "summary": "This cluster is noisy and mostly heterogeneous: the soft pairs span unrelated tokenizer, training, model-loading, and device/runtime bugs. I would not merge any of the proposed soft edges.", + "confidence": 0.68, + "canonical_issue_reason": "issue:45701 is the broadest tokenizer-regression report here and has the strongest signal as an umbrella issue for the tokenizer-related subset of this cluster.", "canonical_pr_reason": null, - "best_issue_reason": "#43452 is the strongest standalone bug report in the set because it is specific, actionable, and likely to remain useful as a reference for nearby loading failures.", + "best_issue_reason": "issue:45701 is the best single issue to anchor the cluster because it is open, broad, and likely to subsume several tokenizer-version regression reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45372", "right": "issue:46170", "accept": false, - "reason": "Both are import-related, but one is about `ReasoningEffort` from `mistral_common` during Gemma processor loading and the other is a Windows/ROCm `torch.distributed` dependency problem. Different failure modes and code paths." - }, - { - "left": "issue:43452", - "right": "issue:43577", - "accept": false, - "reason": "`gguf_file` breaks tokenizer/model loading; the other is a dtype preservation bug in Blip2. Same broad loading context, but not the same underlying bug." + "reason": "Different dependency failures: Gemma processor loading vs Windows/ROCm torch.distributed import/type issues." }, { "left": "issue:44466", "right": "issue:45005", "accept": false, - "reason": "Both mention tied weights in v5, but #44466 is a device-dependent `lm_head.weight` serialization issue while #45005 is a broader translation-model tied-weights complaint. Not concrete enough to merge." + "reason": "Both mention tied weights in v5, but one is a specific lm_head serialization/device bug and the other is a broader translation-model report." }, { - "left": "issue:45542", - "right": "issue:46178", + "left": "issue:43452", + "right": "issue:43577", "accept": false, - "reason": "Both involve backend/optional dependency handling, but one is TensorBoard without TensorFlow and the other is an undefined `tensorflow_text` backend. Different missing-backend bugs." + "reason": "Unrelated problems: gguf_file loading breakage vs BLIP2 dtype propagation." }, { "left": "issue:44493", "right": "issue:46050", "accept": false, - "reason": "Unexpected `position id` keys and broken quantized save/load are unrelated bugs despite both affecting model persistence or outputs." + "reason": "Position-id key mismatches are not the same as sinq quantization save/load breakage." }, { - "left": "issue:43452", - "right": "issue:44843", + "left": "issue:43819", + "right": "issue:44811", "accept": false, - "reason": "Both affect `AutoTokenizer.from_pretrained`, but one is a `gguf_file` break and the other is an offline-mode `model_info()` call in `_patch_mistral_regex`. Distinct root causes." + "reason": "Different model paths and symptoms: DAC latent reconstruction vs Whisper batch decode special-token handling." }, { - "left": "issue:43819", - "right": "issue:44811", + "left": "issue:43452", + "right": "issue:44843", "accept": false, - "reason": "One is a DAC reconstruction mismatch, the other is Whisper decode skipping special tokens. No shared code path." + "reason": "Tokenizer/model loading failures, but one is gguf_file-related and the other is an offline mistral-regex cache lookup bug." }, { "left": "issue:30064", "right": "issue:33666", "accept": false, - "reason": "Image segmentation map processing and Qwen2-VL multi-GPU training are unrelated." + "reason": "Different domains: image processor segmentation-map handling vs Qwen2-VL multi-GPU training." }, { "left": "issue:44936", "right": "issue:45823", "accept": false, - "reason": "Trainer post-train evaluation failure and Gemma4 device-map mismatch are separate runtime issues." + "reason": "Trainer lifecycle bug vs Gemma4 device-map forward mismatch; not the same code path." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Both concern backend detection, but one is TensorFlow/tensorboard setup and the other is tensorflow_text mapping." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "A `max_shard_size` documentation/behavior question is unrelated to `AutoProcessor.from_pretrained` dropping kwargs." + "reason": "save_pretrained shard-size behavior is unrelated to AutoProcessor forwarding kwargs to cached_file." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for missing PyTorch is unrelated to Trainer NoneType broadcast errors." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer/processor regressions, but one is a custom-model processor API break and the other is Kimi-K2.5 codec handling. Different problems." + "reason": "Both are regressions, but one is processor/tokenizer API breakage and the other is Kimi codec/regex handling." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird special-token registration bug is unrelated to MLukeTokenizer AttributeError failures." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer implementations and different failure modes; no shared concrete bug." }, { "left": "issue:28282", - "right": "issue:32090", - "accept": false, - "reason": "Missing PyTorch import and Trainer GPU broadcast `NoneType` are unrelated." - }, - { - "left": "issue:43653", - "right": "issue:44361", + "right": "issue:33357", "accept": false, - "reason": "BigBird special-token registration and MLukeTokenizer task-time AttributeError are different tokenizer bugs." + "reason": "PyTorch import absence and MacOS bus errors on CLIP are unrelated runtime issues." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both hit `apply_chat_template`, but one is batched padding behavior and the other is assistant tool-call messages with empty content. Not the same concrete bug." + "reason": "Both involve apply_chat_template, but the crash conditions differ: batched padding vs tool-call assistant messages." }, { - "left": "issue:44361", - "right": "issue:44442", + "left": "issue:38175", + "right": "issue:43295", "accept": false, - "reason": "MLukeTokenizer AttributeError and FastSpeech2ConformerTokenizer loading failure are unrelated." + "reason": "Siglip2 zero probabilities and processor.tokenizer regression are unrelated issues." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward and Qwen3.5 recurrent-state reset are separate generation bugs." - }, - { - "left": "issue:38175", - "right": "issue:43295", - "accept": false, - "reason": "Siglip2 zero probabilities and a processor/tokenizer regression are unrelated." + "reason": "Chunked generation with compiled forward is not the same as recurrent state reset in Qwen3-MoE cache handling." }, { "left": "issue:28282", - "right": "issue:35141", + "right": "issue:33357", "accept": false, - "reason": "Missing PyTorch and embedding reinitialization after resize are unrelated." + "reason": "Missing PyTorch import and community-CLIP bus error are different failures with different triggers." }, { - "left": "issue:45491", - "right": "issue:46002", + "left": "issue:29159", + "right": "issue:33357", "accept": false, - "reason": "Both mention NaNs, but they affect different models and different mechanisms." + "reason": "Slow/fast tokenizer inconsistency is not the same as the MacOS CLIP bus error." }, { - "left": "issue:41720", - "right": "issue:43906", + "left": "issue:43531", + "right": "issue:43742", "accept": false, - "reason": "Qwen3 device-map CUDA assertion and an isolated reproduction of another issue are not the same bug." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to MobileLLM loading key errors." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer issues, but BigBird special-token registration and missing BOS/EOS on mdeberta-v3-base are separate bugs." + "reason": "Both are tokenizer special-token issues, but they affect different models and different mechanics." }, { "left": "issue:43577", "right": "issue:43697", "accept": false, - "reason": "Blip2 dtype handling and RTDetrV2 output divergence are unrelated." + "reason": "BLIP2 dtype casting and RTDetrV2 output divergence are different bugs." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "MoE auxiliary-loss normalization and T5 apex RMSNorm memory leak are unrelated." + "reason": "Auxiliary-loss normalization and apex RMSNorm memory leak are unrelated." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Offline-mode `model_info()` bug and est-roberta tokenizer loading failure are not the same root cause." + "reason": "Both are tokenizer-loading regressions, but the root causes and affected models differ." }, { "left": "issue:43646", "right": "issue:43749", "accept": false, - "reason": "Custom model initialization breakage in v5 and FSDP CPU RAM efficient loading are different subsystems." + "reason": "Custom model initialization breakage is not the same as FSDP CPU RAM efficient loading." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Aux-loss normalization in MoE models and DeepSeekV3 remote/native divergence are different model correctness issues." + "reason": "Different model families and different failure types: auxiliary-loss math vs DeepSeekV3MoE implementation divergence." }, { - "left": "issue:29159", - "right": "issue:33357", + "left": "issue:43688", + "right": "issue:45701", "accept": false, - "reason": "Slow/fast tokenizer inconsistency and MacOS bus error with a community CLIP model are unrelated." + "reason": "Auxiliary-loss normalization is unrelated to the broad tokenization regression report." }, { - "left": "issue:43688", - "right": "issue:45701", + "left": "issue:43653", + "right": "issue:43927", "accept": false, - "reason": "MoE loss normalization and general tokenization changes across versions are not the same bug." + "reason": "Special-token registration and DiaConfig save/load custom token ID loss are different bugs." } ] }, @@ -6663,7 +6574,9 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ + "issue:28218", "issue:32090", + "issue:33357", "issue:33666", "issue:34567", "issue:36010", @@ -6676,13 +6589,11 @@ "issue:43493", "issue:43504", "issue:43525", - "issue:43531", "issue:43550", "issue:43572", "issue:43577", "issue:43606", "issue:43643", - "issue:43653", "issue:43688", "issue:43720", "issue:43742", @@ -6703,7 +6614,7 @@ "issue:44849", "issue:44913", "issue:44977", - "issue:45357", + "issue:45491", "issue:45588", "issue:45823", "issue:46002", @@ -6712,13 +6623,12 @@ "issue:46209" ], "soft_pairs": [ - "issue:43531|issue:43742", - "issue:43653|issue:43927", - "issue:44464|issue:44977", + "issue:45491|issue:46002", "issue:43550|issue:43927", + "issue:44464|issue:44977", "issue:43643|issue:44561", - "issue:44479|issue:44625", "issue:43550|issue:44361", + "issue:44479|issue:44625", "issue:41720|issue:44464", "issue:44977|issue:46209", "issue:43577|issue:45588", @@ -6732,253 +6642,247 @@ "issue:33666|issue:36010", "issue:43742|issue:44821", "issue:45823|issue:46050", - "issue:43452|issue:44451", "issue:43720|issue:43927", "issue:43606|issue:43927", - "issue:44849|issue:46209", + "issue:43452|issue:44451", "issue:43262|issue:43295", + "issue:44849|issue:46209", "issue:43278|issue:43688", "issue:43994|issue:46192", "issue:43525|issue:43572", "issue:33666|issue:36331", "issue:43720|issue:44361", - "issue:44625|issue:46002", "issue:43756|issue:44913", + "issue:44625|issue:46002", "issue:43493|issue:43825", - "issue:44849|issue:45357" + "issue:28218|issue:33357" ], - "prepared_review_unit_hash": "3abc44c704caff2576a8a1e65be1690cf56ebde2e2d1aa92a3779488b8e33fc5", + "prepared_review_unit_hash": "3fcc6c5a82c56003876caf027180d5ae02392e777fdfc632582c7b20a15799e7", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 34, - "serialized_chars": 37326, - "estimated_input_tokens": 9332, - "estimated_eval_tokens": 18920 + "soft_pair_count": 33, + "serialized_chars": 37049, + "estimated_input_tokens": 9263, + "estimated_eval_tokens": 18782 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft edge here looks like a true duplicate: the items are mostly unrelated bugs/regressions across training, loading, tokenization, multimodal processing, and generation. The cluster is therefore better treated as a noisy similarity bucket than a single duplicate set.", - "confidence": 0.88, - "canonical_issue_reason": "No single canonical issue is safe to pick as the shared bug: the cluster spans multiple unrelated failure modes and model families, so there is no clear duplicate anchor.", + "summary": "This cluster looks like a broad false-positive mix of unrelated issues: tokenizer spacing, trainer errors, model-loading regressions, dtype/device mismatches, and task-specific model bugs. None of the soft edges looks like the same underlying bug or change.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43504 is the cleanest concrete loading-regression report to use as an anchor, though this cluster is not a real duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43504 is the closest thing to a representative hub because it is a concrete loading/config regression and sits near several other load/serialization-themed items, but it still should not be treated as the true canonical duplicate.", + "best_issue_reason": "43504 is the most self-contained and actionable issue here; it has a specific failure mode and is easier to triage than the more model-specific or symptom-only reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43531", - "right": "issue:43742", + "left": "issue:45491", + "right": "issue:46002", "accept": false, - "reason": "Different problems: Qwen3-MoE sliding_window behavior vs a MobileLLM loading key error." + "reason": "Both mention NaNs, but one is a Gemma3 sliding-window GPU bug and the other is a classification-head checkout/init issue; different code paths." }, { - "left": "issue:43653", + "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "Tokenizer special-token registration bug vs config save/load losing custom token IDs; related domain, different root cause." + "reason": "One is a Bamba torch.compile/SDPA failure; the other is a DiaConfig save/load field-loss bug. No shared underlying defect." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is not the same as a Qwen3.5 flash-attention generation failure." - }, - { - "left": "issue:43550", - "right": "issue:43927", - "accept": false, - "reason": "Bamba torch.compile/SDPA failure and config token-ID loss are unrelated code paths." + "reason": "Both are generation regressions, but they involve different models and different causes (compiled forward vs flash-attention)." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "Both involve trust_remote_code, but one is missing fields in AutoConfig and the other is a v5 API removal breakage." + "reason": "Both touch trust_remote_code, but one is missing fields from AutoConfig and the other is a helper removal breaking remote-code models; not the same bug." }, { - "left": "issue:44479", - "right": "issue:44625", + "left": "issue:43550", + "right": "issue:44361", "accept": false, - "reason": "Video-input regression in several VL models vs num_labels propagation in Qwen3.5; different surfaces and fixes." + "reason": "A compile/SDPA model bug versus a tokenizer AttributeError on tasks; unrelated." }, { - "left": "issue:43550", - "right": "issue:44361", + "left": "issue:44479", + "right": "issue:44625", "accept": false, - "reason": "A compile/SDPA attention issue is unrelated to an MLukeTokenizer AttributeError." + "reason": "Both are Qwen-related, but one is a video-input regression and the other is num_labels propagation." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Qwen3 auto device-mapping cuda assert and chunked generation with compiled forward are different bugs." + "reason": "Qwen3 auto-device-map cudaErrorAssert and compiled-forward chunked generation inconsistency are different failures." }, { "left": "issue:44977", "right": "issue:46209", "accept": false, - "reason": "Flash-attention generation bug and FP8/B200 inference failure are different hardware/code-path issues." + "reason": "Qwen3.5 flash-attention generation issues versus FP8 inference on B200; different code paths." }, { "left": "issue:43577", "right": "issue:45588", "accept": false, - "reason": "Blip2 dtype-loading drift is unrelated to flash_attention crashing on s_aux=None." + "reason": "BLIP2 dtype-loading mismatch and flash_attention.py s_aux=None handling are unrelated." }, { "left": "issue:32090", "right": "issue:34567", "accept": false, - "reason": "Trainer gpu_broadcast NoneType error and TrainerState token-count update bug are distinct Trainer issues." + "reason": "Trainer broadcast TypeError versus TrainerState token-counter not updating; different trainer bugs." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "Model dtype loading issue vs MPS attention correctness issue; not the same failure mode." + "reason": "MPS attention correctness and BLIP2 dtype behavior are unrelated." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "BEiT legacy-field load failure is not the same as Bamba torch.compile/SDPA breakage." + "reason": "Legacy-field model loading failure is not the same as the Bamba compile/SDPA regression." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Legacy-field pretrained loading bug and BitNet packed-weight unpacking bug are different root causes." + "reason": "Both are loading regressions, but one is a legacy-field preset load issue and the other is packed-weight unpacking during accelerate loading." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "Model-preset load regression does not match a processor/model input-size mismatch." + "reason": "Model preset loading failure and OmDet processor/model input-size mismatch are different bugs." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Both touch save/load behavior, but one is lm_head serialization and the other is sinq quantized checkpoint handling." + "reason": "Tied-weight serialization depending on device and quantized save/load breakage are not the same code-path problem." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer NoneType broadcast crash is not the same as a multi-GPU training request/problem statement." + "reason": "Trainer NoneType broadcast and Qwen2-VL multi-GPU training are unrelated." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Multi-GPU training question and GenerationMixin import error are unrelated." + "reason": "Multi-GPU training versus GenerationMixin import failure; different subsystems." }, { "left": "issue:43742", "right": "issue:44821", "accept": false, - "reason": "MobileLLM loading key error and AutoImageProcessor-from-URL failure are different loading paths." + "reason": "MobileLLM load key error and AutoImageProcessor-from-URL failure are unrelated." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Gemma4 device-map forward mismatch and sinq save/load corruption are unrelated." - }, - { - "left": "issue:43452", - "right": "issue:44451", - "accept": false, - "reason": "gguf_file breaking tokenizer/model loading is not the same as a specific ScandiBERT load failure." + "reason": "Gemma4 device mismatch with device_map='auto' is not the same as sinq quantized save/load breakage." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "BitNet packed-weight loading and config token-ID persistence are different bug classes." + "reason": "Packed-weight accelerate loading bug versus DiaConfig missing fields; unrelated." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU offload device mismatch and config token-ID loss are unrelated." + "reason": "CPU-offload device mismatch and config field loss are different bugs." }, { - "left": "issue:44849", - "right": "issue:46209", + "left": "issue:43452", + "right": "issue:44451", "accept": false, - "reason": "Qwen3.5 hidden-states bug and FP8 B200 inference failure are different issues." + "reason": "gguf_file/from_pretrained breakage and ScandiBERT load failure are too different to treat as one bug." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Audio chat-template sampling-rate default and processor.tokenizer/image passthrough regression are different processor bugs." + "reason": "Audio apply_chat_template sampling-rate default issue versus processor.tokenizer/image passthrough regression; different APIs." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 hidden-states bug and FP8 inference failure are unrelated." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "Embedding dtype drift during eval and auxiliary-loss normalization are unrelated." + "reason": "BF16-to-FP32 dtype drift and auxiliary-loss normalization are different layers of the stack." }, { "left": "issue:43994", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 incorrect outputs and return_tensors being ignored are different API failures." + "reason": "SigLIP2 output quality issue versus return_tensors being ignored with text_kwargs; not the same defect." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Similar symptom class, but missing pad_token_id in Llama4Config and missing pad_token_idx in StableLmConfig are separate model/config regressions." + "reason": "Both are missing-pad-token config regressions, but they affect different model configs and field names; no shared concrete bug is shown." }, { "left": "issue:33666", "right": "issue:36331", "accept": false, - "reason": "A multi-GPU training topic and a CustomTrainer compute_loss signature regression are not duplicates." + "reason": "Qwen2-VL multi-GPU training and CustomTrainer compute_loss signature mismatch are unrelated." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "BitNet packed-weight loading and MLukeTokenizer task AttributeError are unrelated." + "reason": "BitNet packed-weight loading and MLukeTokenizer task AttributeError have no common code path." }, { - "left": "issue:44625", - "right": "issue:46002", + "left": "issue:43756", + "right": "issue:44913", "accept": false, - "reason": "Qwen3.5 num_labels propagation and NaNs in classification heads are different model-training problems." + "reason": "Smollm3 RoPE layering bug and GPTNeoX rotary_pct reload bug are both config-adjacent, but not the same underlying issue." }, { - "left": "issue:43756", - "right": "issue:44913", + "left": "issue:44625", + "right": "issue:46002", "accept": false, - "reason": "Smollm3 RoPE-layer configuration bug and GPTNeoX rotary_pct reload bug are distinct config issues." + "reason": "num_labels propagation and NaNs in classification heads are unrelated." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 implementation discrepancy and a pipeline error-message regression are unrelated." + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task messaging are unrelated." }, { - "left": "issue:44849", - "right": "issue:45357", + "left": "issue:28218", + "right": "issue:33357", "accept": false, - "reason": "Both are Qwen3.5 regressions, but hidden_states behavior and save_pretrained visual-encoder-key serialization are different bugs." + "reason": "Tokenizer extra-space after added token and MacOS bus error on a CLIP model are unrelated." } ] }, @@ -7002,7 +6906,6 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ - "issue:28218", "issue:33357", "issue:33666", "issue:39692", @@ -7038,11 +6941,13 @@ "issue:44734", "issue:44792", "issue:44843", + "issue:44849", "issue:44964", "issue:44977", "issue:45083", "issue:45200", "issue:45229", + "issue:45357", "issue:45405", "issue:45520", "issue:45584", @@ -7051,245 +6956,251 @@ "issue:45823" ], "soft_pairs": [ - "issue:28218|issue:33357", "issue:44964|issue:45405", + "issue:44849|issue:45357", "issue:39692|issue:43493", "issue:44568|issue:45812", "issue:43540|issue:44977", "issue:43334|issue:43572", "issue:44488|issue:44843", - "issue:43278|issue:43844", "issue:43493|issue:43844", + "issue:43278|issue:43844", "issue:45200|issue:45229", "issue:43756|issue:45083", - "issue:43334|issue:45759", "issue:39692|issue:43295", + "issue:43334|issue:45759", "issue:43257|issue:43299", "issue:43646|issue:44466", "issue:43824|issue:44589", "issue:33357|issue:33666", - "issue:45229|issue:45823", "issue:43504|issue:43606", + "issue:45229|issue:45823", "issue:43927|issue:44361", "issue:44589|issue:45520", "issue:43572|issue:44291", "issue:44464|issue:45584", - "issue:43606|issue:44112", "issue:43278|issue:43701", + "issue:43606|issue:44112", "issue:43504|issue:43653", "issue:43575|issue:45405", "issue:44792|issue:45405", "issue:43994|issue:44493", "issue:44423|issue:44734" ], - "prepared_review_unit_hash": "a4ad9fce9289892764d44806023d4e4c864c37766b959de6c577166ac0857152", + "prepared_review_unit_hash": "1380460df57be53af670159d95aa560322f38a4050a5e5e6ff266196cb904194", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36317, - "estimated_input_tokens": 9080, - "estimated_eval_tokens": 18416 + "serialized_chars": 36911, + "estimated_input_tokens": 9228, + "estimated_eval_tokens": 18712 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is heterogeneous rather than a single duplicate cluster: it mixes unrelated tokenizer, multimodal, config-loading, training, and serve/runtime bugs. There are some thematic overlaps (especially tokenizer/model-loading regressions), but none are close enough to merge as the same underlying issue.", - "confidence": 0.92, - "canonical_issue_reason": "No single canonical issue cleanly represents the whole set because the items span unrelated bugs. If forced to pick the broadest representative, issue 45812 is the most general tokenizer regression, but it does not unify the cluster.", + "summary": "This cluster is mostly heterogeneous: many separate model-specific regressions and config/tokenizer bugs, with only loose topical similarity around Transformers 5.x compatibility and multimodal loading. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45812 is the broadest and most user-impacting issue here (tokenization regression across Granite models) and is the closest thing to a cluster anchor, though the surrounding issues are not true duplicates of it.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45812 is the strongest broad representative among the issues because it describes a wide tokenizer regression affecting multiple Granite models and is more general than the other, mostly model-specific reports.", + "best_issue_reason": "issue:45812 is the strongest global representative because it describes a concrete, high-signal regression with broad model impact and multiple references, unlike the mostly isolated model-specific bugs elsewhere in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:28218", - "right": "issue:33357", + "left": "issue:44964", + "right": "issue:45405", "accept": false, - "reason": "Tokenizer spacing bug vs MacOS bus error in CLIP loading; unrelated symptoms and code paths." + "reason": "Unrelated: model loading failure for Phi-4 multimodal vs an unreleased PEFT version bump." }, { - "left": "issue:44964", - "right": "issue:45405", + "left": "issue:44849", + "right": "issue:45357", "accept": false, - "reason": "One is a Phi-4 multimodal model load failure, the other is a PEFT version pin mismatch; not the same bug." + "reason": "Different bugs: Qwen3.5 hidden-states behavior vs incorrect save_pretrained visual encoder keys." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Both involve SigLIP2, but one is a broken docs example and the other is an implementation discrepancy; different concrete problems." + "reason": "One is a broken documentation/example flow; the other is a SigLIP2 implementation mismatch vs JAX." }, { "left": "issue:44568", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but on different models and different failure modes; not clearly the same underlying bug." + "reason": "Both tokenizer-related, but different regressions and models: mDeBERTa special-token insertion vs Granite token IDs." }, { "left": "issue:43540", "right": "issue:44977", "accept": false, - "reason": "Different models and different failures: video-input processing in Qwen3OmniMoe vs flash-attention generation issues in Qwen3.5." + "reason": "Different failure paths: Qwen3OmniMoe video-processing ValueError vs Qwen3.5 flash-attention generation issue." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Both are missing-token-ID config regressions, but for different model families and different fields; not the same issue." + "reason": "Superficially similar missing config fields, but different models, attributes, and code paths." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "A model load failure for sleng-bert vs an offline-mode regression in AutoTokenizer; unrelated causes." + "reason": "Different subsystems: model loading for sleng-bert vs offline tokenizer patching via model_info()." }, { - "left": "issue:43278", + "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "Embedding dtype drift during eval vs gradient blow-up under ZeRO-3; related training context, but not the same bug." + "reason": "SigLIP2 fidelity mismatch and DeepSpeed gradient inflation are unrelated bugs." }, { - "left": "issue:43493", + "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation mismatch and a DeepSpeed gradient anomaly are distinct problems." + "reason": "Embedding dtype drift in eval vs gradient growth under ZeRO-3 are distinct training issues." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Defaulting mm_token_type_ids to zeros is a config/input fix; the other is a multi-GPU OOM problem." + "reason": "Defaulting mm_token_type_ids to zeros is unrelated to Gemma4 multi-GPU OOM." }, { "left": "issue:43756", "right": "issue:45083", "accept": false, - "reason": "Different model families and different functional areas; no shared concrete code-path bug." + "reason": "Different models and symptoms: RoPE layer count bug vs a helper-length calculation issue." }, { - "left": "issue:43334", - "right": "issue:45759", + "left": "issue:39692", + "right": "issue:43295", "accept": false, - "reason": "Missing pad_token_id and unwrapping text_config are separate multimodal config issues." + "reason": "Doc/example mistakes for SigLIP2 are not the same as the processor.tokenizer regression." }, { - "left": "issue:39692", - "right": "issue:43295", + "left": "issue:43334", + "right": "issue:45759", "accept": false, - "reason": "Both are multimodal-related regressions, but one is a docs/model-processor mismatch and the other is a processor/tokenizer API break." + "reason": "Both involve multimodal config handling, but one is missing pad_token_id and the other is unwrap-text_config behavior." }, { "left": "issue:43257", "right": "issue:43299", "accept": false, - "reason": "Both mention Qwen3 MoE, but one is missing weight conversion under accelerate+deepspeed and the other is a model-loading regression in dev." + "reason": "Related family, but different concrete problems: loading/weight conversion under accelerate+deepspeed vs Qwen3VL Moe loading failure." }, { "left": "issue:43646", "right": "issue:44466", "accept": false, - "reason": "Custom model initialization breakage and inconsistent lm_head serialization are different v5 regressions." + "reason": "Custom model initialization breakage and device-dependent lm_head serialization are different regressions." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "ImportError for a missing model class vs a Float8 storage TypeError; unrelated." + "reason": "Importing a missing model class and a Float8 storage lookup failure are unrelated." }, { "left": "issue:33357", "right": "issue:33666", "accept": false, - "reason": "MacOS bus error with CLIP and Qwen2-VL multi-GPU training are unrelated failures." + "reason": "CLIP bus error on macOS and Qwen2-VL multi-GPU training are completely different issues." }, { - "left": "issue:45229", - "right": "issue:45823", + "left": "issue:43504", + "right": "issue:43606", "accept": false, - "reason": "OOM during Gemma4 inference vs device mismatch with device_map=auto; different bugs." + "reason": "Different models and code paths: BEiT legacy-field loading vs bark-small CPU offload device mismatch." }, { - "left": "issue:43504", - "right": "issue:43606", + "left": "issue:45229", + "right": "issue:45823", "accept": false, - "reason": "Legacy-field model loading issue vs CPU offload device mismatch; not the same underlying defect." + "reason": "Same model family, but one is OOM during inference and the other is a device-placement mismatch during forward." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "Config save/load token ID loss and MLukeTokenizer AttributeError are distinct tokenizer issues." + "reason": "Both tokenizer/config related, but the underlying failures and models differ." }, { "left": "issue:44589", "right": "issue:45520", "accept": false, - "reason": "Different storage/import failures with no shared code-path or symptom." + "reason": "Float8 storage type resolution and Python 3.13 flash_attn import metadata are unrelated." }, { "left": "issue:43572", "right": "issue:44291", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig vs unexpected _is_hf_initialized argument in init_empty_weights; separate regressions." + "reason": "Both are Transformers 5 compatibility problems, but one is a config field rename and the other is an unexpected constructor argument." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward and Whisper empty-transcription failure are unrelated." + "reason": "Different generation bugs: compiled-forward chunking inconsistency vs empty-transcription handling in Whisper." }, { - "left": "issue:43606", - "right": "issue:44112", + "left": "issue:43278", + "right": "issue:43701", "accept": false, - "reason": "A runtime CPU offload device mismatch vs a stale CI test; not the same bug." + "reason": "Embedding dtype drift and resume_from_checkpoint key mismatch do not share the same root cause." }, { - "left": "issue:43278", - "right": "issue:43701", + "left": "issue:43606", + "right": "issue:44112", "accept": false, - "reason": "Different failure classes: dtype change during eval vs resume_from_checkpoint key mismatch." + "reason": "A real runtime device mismatch vs a stale CI test are not the same underlying bug." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "Model preset loading with a legacy field vs tokenizer special-token registration bug; different components and symptoms." + "reason": "BEiT preset loading and BigBirdTokenizer special-token registration are unrelated." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "TP OOM on a large Qwen2 model and an unreleased PEFT version pin are unrelated." + "reason": "TP OOM for a large Qwen model is unrelated to the PEFT version bump issue." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "A Janus test failure and a dependency version pin issue do not share the same underlying bug." + "reason": "A failed Janus generation test and an unreleased dependency version bump are unrelated." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 wrong outputs and a broad position-id warning are not the same specific defect." + "reason": "SigLIP2 producing bad outputs and widespread unexpected position-id keys are distinct regressions." }, { "left": "issue:44423", "right": "issue:44734", "accept": false, - "reason": "Both are serve-related, but one is a multimodal .to() crash and the other is KV-cache indexing; different code paths." + "reason": "Both are serve/runtime crashes, but one is a multimodal .to() type error and the other is KV-cache indexing; different code paths." + }, + { + "left": "issue:44843", + "right": "issue:45357", + "accept": false, + "reason": "Offline tokenizer patching and incorrect visual-encoder save keys are separate issues." } ] }, @@ -7364,22 +7275,22 @@ ], "soft_pairs": [ "issue:43873|issue:46050", - "issue:45406|issue:46121", "issue:29942|issue:30333", + "issue:45406|issue:46121", "issue:45588|issue:46050", "issue:42490|issue:43696", "issue:43844|issue:44964", - "issue:29159|issue:32090", "issue:28218|issue:34567", + "issue:29159|issue:32090", "issue:44521|issue:45290", "issue:43582|issue:44977", "issue:44554|issue:45446", "issue:45084|issue:45520", - "issue:43540|issue:45200", "issue:15354|issue:38175", + "issue:43540|issue:45200", "issue:45072|issue:45198", - "issue:45200|issue:45823", "issue:37428|issue:43957", + "issue:45200|issue:45823", "issue:43232|issue:44849", "issue:44186|issue:46192", "issue:43493|issue:43701", @@ -7393,14 +7304,14 @@ "issue:37428|issue:43824", "issue:39692|issue:46192" ], - "prepared_review_unit_hash": "b05542b7f5a895e5f50ccf41ac66bdeb2f1dfd5c462538a30168d3936ae02818", + "prepared_review_unit_hash": "89e2a9f82e740b883b1aeb94fb5cbfc79e77a143ae1a914e384ce7cc22901f06", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, @@ -7415,186 +7326,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These issues are broadly unrelated; they span tokenizer behavior, multimodal processors, generation, loading, and backend-specific runtime errors. None of the soft pairs looks like the same underlying bug/change, so I rejected all of them.", - "confidence": 0.93, - "canonical_issue_reason": "issue:44521 is the strongest standalone representative: it is open, has multiple inbound references, and describes a concrete, reproducible apply_chat_template failure in a multimodal path.", + "summary": "This cluster is highly heterogeneous; the soft edges are mostly title-level similarities across unrelated bugs, so none of the candidate pairs look like true duplicates or mergeable PR-equivalents.", + "confidence": 0.97, + "canonical_issue_reason": "issue:29159 is the broadest, most representative issue in the set (generic tokenizer inconsistency), but the cluster is too mixed for a strong canonical duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 has the best combination of specificity, reproducibility, and active discussion among the listed issues.", + "best_issue_reason": "issue:29159 is the best cluster anchor because it is general and closest to the tokenizer-related subset, even though the overall cluster does not form a single bug family.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Both involve quantization, but one is about offloading behavior and the other about save/load breakage; different bugs." + "reason": "Both mention quantization, but one is about offloading behavior and the other about persistence after save/load; different bugs." }, { - "left": "issue:45406", - "right": "issue:46121", + "left": "issue:29942", + "right": "issue:30333", "accept": false, - "reason": "Gemma4 processor attribute regression vs rope-parameter list typing; unrelated failure modes." + "reason": "Flash Attention test failures and MLflow reporting on failed jobs are unrelated subsystems and failure modes." }, { - "left": "issue:29942", - "right": "issue:30333", + "left": "issue:45406", + "right": "issue:46121", "accept": false, - "reason": "Flash Attention test failures and MLflow job reporting are unrelated subsystems." + "reason": "Gemma4Processor missing _tokenizer is unrelated to rope-parameter validation TypeError." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "Flash-attention sinkless-model crash vs sinq save/load corruption; not the same code-path problem." + "reason": "One is a flash-attention integration AttributeError; the other is quantized-model save/load breakage." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Local-path import warning and CUDA OOM are entirely different issues." + "reason": "Local-path import warnings and CUDA OOM on GPT-oss-20b are clearly different problems." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient growth vs Phi-4 multimodal loading regression; no shared bug." + "reason": "ZeRO-3 gradient growth and Phi-4 multimodal load failure are not the same code-path bug." }, { - "left": "issue:29159", - "right": "issue:32090", + "left": "issue:28218", + "right": "issue:34567", "accept": false, - "reason": "Tokenizer slow/fast inconsistency and Trainer GPU broadcast NoneType error are unrelated." + "reason": "Tokenizer spacing after added tokens is unrelated to TrainerState token-count tracking." }, { - "left": "issue:28218", - "right": "issue:34567", + "left": "issue:29159", + "right": "issue:32090", "accept": false, - "reason": "Added-token spacing bug vs TrainerState token counter not updating; different functionality." + "reason": "Slow/fast tokenizer inconsistency is unrelated to the Trainer _gpu_broadcast_one NoneType error." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both touch apply_chat_template, but one is all-zero assistant masks for multimodal inputs while the other is a crash on tool-call assistant messages; too distinct to merge." + "reason": "Zero assistant masks in multimodal templates and crashes on tool-call messages are different chat-template bugs." }, { "left": "issue:43582", "right": "issue:44977", "accept": false, - "reason": "AppleSilicon allocator TypeError and Qwen3.5 flash-attention generation bug are unrelated." + "reason": "Apple Silicon allocator TypeError and Qwen3.5 flash-attention generation issues do not share a concrete bug." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness issue and a PyTorch version check bug in flex_attention are separate defects." + "reason": "MPS attention correctness and a PyTorch version check/import bug are unrelated." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Template-node compilation error and flash_attn import KeyError on Python 3.13 do not match." + "reason": "Template-node compilation failure and Python 3.13 flash_attn KeyError are different import/compile paths." }, { - "left": "issue:43540", - "right": "issue:45200", + "left": "issue:15354", + "right": "issue:38175", "accept": false, - "reason": "Different Gemma/Qwen-style multimodal issues with different symptoms and model paths." + "reason": "torch.jit.script GeneratorExp export failure and unexpected zero probabilities in SigLIP2 are unrelated." }, { - "left": "issue:15354", - "right": "issue:38175", + "left": "issue:43540", + "right": "issue:45200", "accept": false, - "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero-probabilities are unrelated." + "reason": "Qwen3OmniMoe video-processing ValueError and Gemma4 mm_token_type_ids behavior are separate model-specific issues." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "bfloat16 dtype mismatches in inference and Wav2Vec2 save/tokenization failure are different bugs." + "reason": "dtype mismatch in inference and Wav2Vec2 save/tokenization failures are not the same bug." }, { - "left": "issue:45200", - "right": "issue:45823", + "left": "issue:37428", + "right": "issue:43957", "accept": false, - "reason": "Both are Gemma4-related, but one is about default token type ids and the other about device placement during forward; not the same issue." + "reason": "Flash-attention utility import error and meta-device model loading breakage are distinct loading issues." }, { - "left": "issue:37428", - "right": "issue:43957", + "left": "issue:45200", + "right": "issue:45823", "accept": false, - "reason": "Flash-attention utility import error vs meta-device model-loading regression are separate loading problems." + "reason": "Gemma4 defaulting mm_token_type_ids and PLE device mismatch under device_map=auto are different failures." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwargs sync bug and Qwen3.5 hidden-state bug are different code paths." + "reason": "Generation kwargs after sync_gpus and Qwen3.5 output_hidden_states behavior are separate generation bugs." }, { "left": "issue:44186", "right": "issue:46192", "accept": false, - "reason": "LayoutLMv2 NER/padding crashes vs return_tensors being ignored in chat/template kwargs are unrelated tokenizer bugs." + "reason": "LayoutLMv2Tokenizer crashes on NER/padding and return_tensors being ignored are unrelated tokenizer API issues." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 implementation discrepancy and resume_from_checkpoint key mismatch are not the same defect." + "reason": "SigLIP2 implementation discrepancy and resume_from_checkpoint key mismatch are not the same code path." }, { "left": "issue:45381", "right": "issue:46209", "accept": false, - "reason": "Qwen2.5-VL video vision_position_ids bug and FP8 inference failure on B200 are unrelated." + "reason": "Qwen2.5-VL video position IDs and FP8 inference on B200 are different model/hardware problems." }, { "left": "issue:45823", "right": "issue:46192", "accept": false, - "reason": "Gemma4 device mismatch and return_tensors handling are different issues." + "reason": "Gemma4 device placement and return_tensors handling are unrelated regressions." }, { "left": "issue:43295", "right": "issue:43450", "accept": false, - "reason": "Processor/tokenizer regression and batched video-shape bug affect different paths and models." + "reason": "Gemma4Processor _tokenizer regression and batched video shape handling are different processor issues." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Custom processor.tokenizer regression vs OmDet-Turbo resolution mismatch; not mergeable as one bug." + "reason": "Same as above: processor tokenizer access vs OmDet-Turbo image-size mismatch are unrelated." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "StableLmConfig pad-token regression and a physical-intelligence model load failure are distinct." + "reason": "StableLmConfig missing pad_token_idx and transformers>=5.1.0 loading fast are different versioning/loading failures." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "Attention correctness on MPS vs Kimi-K2.5 tokenizer codec regression are unrelated." + "reason": "MPS attention correctness bug and Kimi-K2.5 tokenizer regression are unrelated." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "Config missing pad_token_idx and meta-device loading breakage are different compatibility issues." + "reason": "Missing pad_token_idx and meta-device loading failures do not describe the same defect." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "Different import errors for different symbols; no common underlying change." + "reason": "A missing flash-attention import symbol and a missing Qwen model class import are distinct import errors." }, { "left": "issue:39692", "right": "issue:46192", "accept": false, - "reason": "SigLIP2 docs/model-processor mismatch and return_tensors handling are unrelated." + "reason": "SigLIP2 documentation/example errors are unrelated to return_tensors being ignored when text_kwargs is passed." } ] }, @@ -7677,229 +7588,229 @@ "issue:43334|issue:43525", "issue:44792|issue:45563", "issue:43824|issue:43975", + "issue:43450|issue:43575", "issue:41720|issue:46209", "issue:42915|issue:43824", - "issue:43450|issue:43575", "issue:43388|issue:43526", "issue:42490|issue:43493", "issue:43278|issue:43856", "issue:28218|issue:35141", - "issue:38175|issue:43493", "issue:43653|issue:44112", - "issue:43975|issue:45520", + "issue:38175|issue:43493", "issue:43526|issue:46192", + "issue:43975|issue:45520", "issue:44442|issue:45198", - "issue:45520|issue:46170", "issue:38175|issue:45412", + "issue:45520|issue:46170", "issue:44514|issue:44521", - "issue:44206|issue:45584", "issue:43262|issue:43450", + "issue:44206|issue:45584", "issue:43827|issue:43901", "issue:43819|issue:44112", "issue:44610|issue:44857" ], - "prepared_review_unit_hash": "c1d3ae045b663761b81e90587a376b8801c3475e10e6aed08e1b4c15054312a3", + "prepared_review_unit_hash": "93acfbd16f6cb607b323d2e8bfe4260f7147f40c15fbd1022929cdf37a13f6d0", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36619, - "estimated_input_tokens": 9155, - "estimated_eval_tokens": 18566 + "serialized_chars": 36622, + "estimated_input_tokens": 9156, + "estimated_eval_tokens": 18568 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is heterogeneous: tokenizer, multimodal processor, model-loading, training, and docs regressions all appear mixed together. I do not see any soft pair that looks like the same underlying bug/change, so I reject all soft edges.", - "confidence": 0.88, - "canonical_issue_reason": "issue:44521 is the strongest representative bug report in the set: it targets a single concrete runtime failure in `apply_chat_template` for multimodal inputs, has clear API scope, and enough discussion to be triage-friendly.", + "summary": "This cluster is heterogeneous; none of the soft pairs look like true duplicates, so I rejected all of them. The only weakly central issue is the import/backend failure around `flash_attn` (45520), but even that is just the best representative of a noisy cluster, not a real duplicate hub.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 45520 is the most connected item in the set and is a concrete, reproducible bug report; it is the closest thing to a cluster anchor, though the surrounding issues are still unrelated.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 is the best issue to anchor on because it is specific, reproducible, and clearly describes a user-facing regression in a named code path.", + "best_issue_reason": "45520 is the clearest and most central issue to use as a representative for this noisy cluster, but it should not be treated as a true duplicate canonical without stronger evidence.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Different bugs in different model families: Gemma 4 token-type defaults vs Qwen2.5-VL rope-index temporal scaling." + "reason": "Different bugs in different multimodal paths: Gemma 4 text-only token-type defaults vs Qwen2.5-VL temporal position scaling." }, { "left": "issue:45520", "right": "issue:46178", "accept": false, - "reason": "Both are import/backend problems, but one is a `flash_attn` KeyError on Python 3.13 and the other is a TensorFlow Text backend mapping error; not the same failure." + "reason": "Both are import/backend-related, but they hit different backend symbols and different failure modes; not the same underlying bug." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Different code paths and symptoms: processor/tokenizer image passing regression versus quantization offloading not behaving as expected." + "reason": "One is a processor API regression, the other is quantization/offloading behavior; too far apart." }, { "left": "issue:43550", "right": "issue:43844", "accept": false, - "reason": "Both mention training/runtime instability, but one is a Bamba SDPA `torch.compile` failure and the other is gradient explosion with DeepSpeed ZeRO-3." + "reason": "Different code paths and symptoms: torch.compile with SDPA vs ZeRO-3 gradient blow-up." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "One is a Perceiver non-default-resolution forward-path failure; the other is RT-DETR memory not being released after deletion." + "reason": "Unrelated issues: Perceiver resize/interpolation failure vs RT-DETR memory not being freed." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Both involve weights/serialization behavior, but one is inconsistent `lm_head.weight` saving and the other is LoRA merge collapse with extended vocabulary." + "reason": "Both mention tied weights, but one is device-dependent serialization of `lm_head.weight` and the other is LoRA merge collapse with extended vocab; not the same bug." }, { "left": "issue:43334", "right": "issue:43525", "accept": false, - "reason": "Same broad symptom (`pad_token_id` missing), but different model configs and likely different fixes; not clearly one duplicate bug." + "reason": "Same broad symptom (`pad_token_id` missing), but for different model configs and likely different fixes." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "The first is a Janus image-generation test failure; the second is a stale paged-generate warning. Different issues." + "reason": "Different failures: a Janus image-generation test vs a stale paged-generate warning." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "ImportError for `Qwen2_5_VLForConditionalGeneration` is unrelated to a DeepSeek detokenization regression." + "reason": "Import error for a missing model class vs detokenization regression in DeepSeek Coder; no shared code-path evidence." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Video processor batching shape bug vs tensor-parallel OOM on a large model; unrelated." }, { "left": "issue:41720", "right": "issue:46209", "accept": false, - "reason": "Both are CUDA/FP8-related runtime failures, but one is Qwen3 auto device mapping on A800 and the other is FP8 inference on B200; different problem statements." + "reason": "Both are hardware-specific inference failures, but on different models and different accelerator constraints; not mergeable as one fix." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Qwen3Moe FP8 failure is unrelated to a missing Qwen2.5-VL import symbol." - }, - { - "left": "issue:43450", - "right": "issue:43575", - "accept": false, - "reason": "Video processor batch-shape bug versus tensor-parallel OOM on Qwen2-57B; not the same underlying defect." + "reason": "Qwen3MoE FP8 failure and Qwen2.5-VL import error are unrelated." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Different label-handling bugs in different components: metric gathering drops tuple labels vs BeitImageProcessorFast returning only one label." + "reason": "Both concern labels, but one is metric gathering dropping tuple elements and the other is BeitImageProcessorFast reducing labels incorrectly; different behaviors." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "A trailing-slash local-model import warning is not the same as SigLIP2 implementation mismatch versus JAX." + "reason": "Path-trailing-slash import warning is unrelated to SigLIP2 implementation discrepancies." }, { "left": "issue:43278", "right": "issue:43856", "accept": false, - "reason": "Embedding dtype drift in evaluate is not the same as memory inefficiency in Qwen3 MoE training." + "reason": "Training/eval dtype mismatch vs MoE memory inefficiency; different defects." }, { "left": "issue:28218", "right": "issue:35141", "accept": false, - "reason": "Tokenizer extra-space detokenization bug versus embedding reinitialization after `resize_token_embeddings`; different subsystems." - }, - { - "left": "issue:38175", - "right": "issue:43493", - "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 and HF/JAX discrepancy are related only thematically; they describe different concrete bugs." + "reason": "Tokenizer spacing after added tokens and embedding reinitialization after resize are different issues." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "Both are CI/test-related, but one is a BigBirdTokenizer special-token registration bug and the other is a stale GraniteSpeech device-override test." + "reason": "BigBirdTokenizer special-token registration bug vs GraniteSpeech CI test staleness; not the same bug." }, { - "left": "issue:43975", - "right": "issue:45520", + "left": "issue:38175", + "right": "issue:43493", "accept": false, - "reason": "DeepSeek detokenization in v5 is unrelated to a Python 3.13 `flash_attn` import failure." + "reason": "Both mention SigLIP2, but one is a specific zero-probability symptom and the other is a broader HF-vs-JAX discrepancy; insufficient evidence of one underlying bug." }, { "left": "issue:43526", "right": "issue:46192", "accept": false, - "reason": "One is BeitImageProcessorFast label reduction; the other is `return_tensors` being ignored when `text_kwargs` is passed." + "reason": "Label reduction bug and `return_tensors` being ignored are unrelated processor issues." }, { - "left": "issue:44442", - "right": "issue:45198", + "left": "issue:43975", + "right": "issue:45520", "accept": false, - "reason": "Different tokenizer failures: AutoTokenizer cannot load a tokenizer class, while Wav2Vec2 save_pretrained/tokenization breaks for a specific model." + "reason": "Detokenization regression in a model tokenizer is unrelated to the `flash_attn` import_utils KeyError." }, { - "left": "issue:45520", - "right": "issue:46170", + "left": "issue:44442", + "right": "issue:45198", "accept": false, - "reason": "Both involve imports, but one is a Python 3.13 backend lookup error and the other is Windows/ROCm `torch.distributed` dependency handling." + "reason": "AutoTokenizer loading failure for one tokenizer and save_pretrained/tokenization failure for another are separate tokenizer bugs." }, { "left": "issue:38175", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 numerical mismatch is unrelated to RT-DETR model memory not being released." + "reason": "SigLIP2 zero-probabilities bug vs RT-DETR memory cleanup bug; unrelated." }, { - "left": "issue:44514", - "right": "issue:44521", + "left": "issue:45520", + "right": "issue:46170", "accept": false, - "reason": "Same processor family, but different concrete failures: batched `apply_chat_template` crash with `padding=False` vs all-zero assistant masks for multimodal inputs." + "reason": "Different dependency/import problems: `flash_attn` backend mapping vs torch.distributed on Windows/ROCm." }, { - "left": "issue:44206", - "right": "issue:45584", + "left": "issue:44514", + "right": "issue:44521", "accept": false, - "reason": "LASR feature-extractor center-arg regression is unrelated to Whisper empty-transcription failure after `align_special_tokens`." + "reason": "Both touch `apply_chat_template`, but one crashes on batched padding=False input and the other returns zero assistant masks; different concrete failures." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio chat-template sampling-rate default bug versus video processor batched-shape bug; different processors and symptoms." + "reason": "Audio processor sample-rate defaulting and batched video processor shapes are unrelated." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Feature extractor `center` arg crash and Whisper empty-transcription generation failure are different code paths." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs-related, but one references removed `pipeline()` usage and the other mentions `return_all_scores`; not the same change." + "reason": "Both are docs regressions around pipeline deprecations, but they refer to different pages and API details; too broad to merge as one fix." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "DAC latent reconstruction mismatch is unrelated to a GraniteSpeech stale device-override CI test." + "reason": "DAC latent/forward mismatch is unrelated to a GraniteSpeech CI test issue." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "OmDet-Turbo processor output size mismatch and LwDetrImageLoss float16/CUDA crash are different bugs in different components." + "reason": "Different model families and failures: OmDet-Turbo input-size mismatch vs LwDetr AMP/CUDA crash." } ] }, @@ -7978,11 +7889,11 @@ "issue:43335|issue:43550", "issue:43606|issue:45083", "issue:44857|issue:45593", - "issue:44448|issue:44743", "issue:29942|issue:32090", + "issue:44448|issue:44743", "issue:45127|issue:45245", - "issue:45290|issue:45381", "issue:37428|issue:44351", + "issue:45290|issue:45381", "issue:45200|issue:45588", "issue:43278|issue:43825", "issue:43335|issue:43653", @@ -7996,8 +7907,8 @@ "issue:29127|issue:32090", "issue:43335|issue:45083", "issue:39692|issue:43526", - "issue:45310|issue:45478", "issue:39692|issue:44493", + "issue:45310|issue:45478", "issue:43720|issue:44112", "issue:43335|issue:44112", "issue:43335|issue:45072", @@ -8008,20 +7919,20 @@ "issue:43606|issue:44857", "issue:34689|issue:35141" ], - "prepared_review_unit_hash": "b4f7f39a281a2d3ab4d783bf7c3db5e9843708a21ffb1fc23c800a9f23fb5d61", + "prepared_review_unit_hash": "3ed5a871e59065f61a6c0c8dbbb4af2b1be0a50e6163bfc081a226e5b14d244d", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 38093, + "serialized_chars": 38095, "estimated_input_tokens": 9524, "estimated_eval_tokens": 19304 }, @@ -8030,54 +7941,54 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. The only clear duplicate pair is the two identical \"Qwen3.5 Moe from_pretrained error\" issues (45310/45478); the rest are different bugs/regressions in unrelated paths.", + "summary": "These items are mostly unrelated issues with only one clear duplicate pair: the Qwen3.5 Moe from_pretrained regression (#45310 and #45478). All other soft edges look like different bugs or different code paths and should be rejected.", "confidence": 0.94, - "canonical_issue_reason": "issue:45310 is the best anchor because it is the earlier report of the only clear duplicate subgroup, and issue:45478 is an exact title match on the same Qwen3.5 MoE loading failure.", + "canonical_issue_reason": "issue:45310 is the earlier, fuller report of the only clear duplicate pair and cleanly captures the Qwen3.5 Moe from_pretrained regression.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45310 is the strongest representative issue in this cluster because it has the clearest duplicate evidence and best captures the only shared underlying problem here.", + "best_issue_reason": "issue:45310 is the best issue anchor because it is the strongest true duplicate target in the set; the rest are heterogeneous and do not form a coherent duplicate cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "Different failures: DAC latent reconstruction mismatch vs a helper output-length calculation bug." + "reason": "Different model subsystems and failure modes: DAC latents/STE mismatch vs qwen3_omni_moe feature length helper behavior." }, { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "Both mention loading/version problems, but one is a missing internal import and the other is a broad Torch compatibility complaint; not the same bug." + "reason": "One is an import error for a missing symbol; the other is a broad torch version compatibility complaint." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "Unrelated code paths: SwitchTransformers sparse-layer config vs Bamba torch.compile/SDPA failure." + "reason": "SwitchTransformers sparse-layer config bug vs Bamba torch.compile/SDPA failure; unrelated code paths." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU offload device-mismatch in Bark is unrelated to qwen3_omni_moe feature-length behavior." + "reason": "CPU offload device mismatch in bark-small vs a qwen3_omni_moe helper-function behavior issue." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "Different models and symptoms: AMP/CUDA crash in LwDetrImageLoss vs missing auxiliary losses in D-FINE." + "reason": "Float16 AMP crash in LwDetrImageLoss is unrelated to D-FINE missing auxiliary losses." }, { - "left": "issue:44448", - "right": "issue:44743", + "left": "issue:29942", + "right": "issue:32090", "accept": false, - "reason": "Pegasus v4/v5 output drift is not the same as recurrent state reset in modular_qwen3_5 cache handling." + "reason": "Flash Attention 2 test failures vs Trainer/_gpu_broadcast_one NoneType error are different bugs." }, { - "left": "issue:29942", - "right": "issue:32090", + "left": "issue:44448", + "right": "issue:44743", "accept": false, - "reason": "Flash Attention 2 test failures and NoneType broadcast errors in Trainer are unrelated bugs." + "reason": "Pegasus output regression across versions vs recurrent state reset in qwen3 modular code; different mechanisms." }, { "left": "issue:45127", @@ -8086,166 +7997,160 @@ "reason": "LoRA/extended-vocab collapse is unrelated to the 2^24 category limit error." }, { - "left": "issue:45290", - "right": "issue:45381", + "left": "issue:37428", + "right": "issue:44351", "accept": false, - "reason": "Chat-template tool-call crash and Qwen2.5-VL video position-id bug are different features and failure modes." + "reason": "Both are import-related, but they involve different missing symbols and different breakages." }, { - "left": "issue:37428", - "right": "issue:44351", + "left": "issue:45290", + "right": "issue:45381", "accept": false, - "reason": "Both are import errors, but they reference different missing symbols and likely different regression causes." + "reason": "Chat template tool-call crash vs qwen2.5-vl video vision_position_ids mismatch; not the same bug." }, { "left": "issue:45200", "right": "issue:45588", "accept": false, - "reason": "Gemma4 mm_token_type_ids handling is unrelated to flash_attention.py crashing on s_aux=None." + "reason": "Gemma4 mm_token_type_ids defaulting issue is unrelated to flash_attention s_aux=None crashes." }, { "left": "issue:43278", "right": "issue:43825", "accept": false, - "reason": "Embedding dtype drift during evaluate is unrelated to a pipeline translation-task error message regression." + "reason": "Embedding dtype drift during eval vs pipeline translation-task warning; different API surfaces." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and BigBirdTokenizer special-token decoding are unrelated." + "reason": "Sparse layer misconfiguration in SwitchTransformers vs BigBirdTokenizer special-token registration problem." }, { "left": "issue:43611", "right": "issue:43931", "accept": false, - "reason": "Base-model-prefix loading regression and Qwen3-VL weight-shape mismatch are different model-loading failures." + "reason": "Base model prefix loading regression is unrelated to Qwen3-VL weight-shape mismatch." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Unsupported center arg in LasrFeatureExtractor is unrelated to Gemma4 PLE device placement issues." + "reason": "LaserFeatureExtractor center-arg regression vs Gemma4 PLE device mismatch; different failures." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "torch.compile with SDPA in Bamba is unrelated to the GraniteSpeech stale device override test." + "reason": "Bamba torch.compile SDPA bug vs GraniteSpeech stale device-override CI failure are unrelated." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "DAC latent forward mismatch is not the same as unexpected position-id keys across many models." + "reason": "DAC forward/from_latents mismatch is not the same as unexpected position-id key warnings." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 IndentationError in DebertaV2 import is unrelated to qwen3_omni_moe feature-length helper behavior." + "reason": "Python 3.13 IndentationError in DebertaV2 import is unrelated to qwen3_omni_moe output-length helper behavior." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "SwitchTransformers sparse-layer bug and Beit legacy-field loading failure are unrelated." + "reason": "SwitchTransformers sparse-layer creation bug vs Beit semantic-segmentation preset loading issue; different models and causes." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Both involve attention/offload-adjacent code, but one is an AttributeError on s_aux=None and the other is CPU-offload incompatibility in a rotary embedding." + "reason": "Both touch flash/offload behavior, but one is an AttributeError on s_aux=None and the other is CPU offload incompatibility in rotary embedding." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 box-content validation message and NoneType GPU broadcast are unrelated trainer/model bugs." + "reason": "LayoutLMv3 error-message clarity issue vs trainer NoneType broadcast crash; different underlying bugs." }, { "left": "issue:43335", - "right": "issue:45083", + "right": "issue:45072", "accept": false, - "reason": "SwitchTransformers sparse-layer config and qwen3_omni_moe output-length helper are different subsystems." + "reason": "SwitchTransformers sparse-layer bug vs dtype mismatches in bfloat16 inference; not the same defect." }, { "left": "issue:39692", "right": "issue:43526", "accept": false, - "reason": "SigLIP2 documentation example errors and BeitImageProcessorFast reduce_labels behavior are unrelated." - }, - { - "left": "issue:45310", - "right": "issue:45478", - "accept": true, - "reason": "Exact same Qwen3.5 Moe from_pretrained regression; titles and failure description match closely, so these are duplicates." + "reason": "SigLIP2 docs/example errors are unrelated to BeitImageProcessorFast reduce_labels returning the wrong labels." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "Doc-example/model-processor mismatch is unrelated to widespread unexpected position-id keys." + "reason": "Documentation/example failures vs position-id key regression; different artifacts and code paths." }, { - "left": "issue:43720", - "right": "issue:44112", - "accept": false, - "reason": "BitNet packed-weight unpacking during accelerate loading is unrelated to the GraniteSpeech CI device-override failure." + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same title, same model family, same from_pretrained error; clear duplicate reports of the same regression." }, { - "left": "issue:43335", + "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "Sparse-layer config bug and stale device override test do not describe the same underlying problem." + "reason": "BitNet packed-weight unpacking failure is unrelated to GraniteSpeech CI device-override staleness." }, { "left": "issue:43335", - "right": "issue:45072", + "right": "issue:44112", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and bfloat16 dtype mismatches in inference are different defects." + "reason": "Different areas entirely: SwitchTransformers config vs GraniteSpeech CI." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Model output drift in Pegasus and tokenizer errors in Deepseek are unrelated." + "reason": "Pegasus v4/v5 output change is not the same as DeepSeek tokenizer tokenization regression." }, { "left": "issue:44360", "right": "issue:44561", "accept": false, - "reason": "DSA indexer missing ReLU is unrelated to the removal of is_torch_fx_available breaking trust_remote_code models." + "reason": "DSA indexer lacking ReLU vs removal of is_torch_fx_available; unrelated breakages." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Local-path trailing-slash import warning is unrelated to GLM-4.7-Flash unit-test load failures." + "reason": "Local-model path import warning vs failing to load GLM-4.7-Flash in unit tests; different issues." }, { "left": "issue:45920", "right": "issue:46192", "accept": false, - "reason": "AutoTokenizer wrong IDs and return_tensors being ignored are both tokenizer-related, but they are distinct behaviors and not the same bug." + "reason": "AutoTokenizer wrong token IDs across models vs return_tensors being ignored when text_kwargs is passed; similar API area but different bugs." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "CPU offload device mismatch in Bark and float16 AMP crash in LwDetrImageLoss are unrelated." + "reason": "CPU offload device mismatch in Bark-small vs float16 AMP crash in LwDetrImageLoss; not the same code path." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Llama 3.2 vision model loading break and output embedding reinitialization after resize are different issues." + "reason": "Model loading regression for Llama 3.2 Vision is unrelated to embedding reinitialization after resizing." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and conservative overall. The canonical pair choice is supported by the exact title match between issue:45310 and issue:45478, and the rest of the soft-pair rejections stay appropriately cautious given the differing titles/symptoms in the packet. The summary does not overclaim beyond the evidence." + "feedback": "The assessment is grounded in the packet and stays conservative on soft-edge pairs. The only clearly supported duplicate pair is issue:45310 with issue:45478, which share the same title and model-specific from_pretrained regression. The rest of the pair rejections are consistent with the titles/bodies and do not overstate the evidence." }, "evaluator_used": true, "retried": false, @@ -8325,42 +8230,42 @@ "issue:28218|issue:32090", "issue:43701|issue:45083", "issue:32090|issue:33357", - "issue:43550|issue:45083", "issue:29942|issue:35141", + "issue:43550|issue:45083", "issue:43723|issue:44554", "issue:44561|issue:44779", "issue:37428|issue:44987", - "issue:43873|issue:45229", "issue:43927|issue:45698", + "issue:43873|issue:45229", "issue:44462|issue:44561", "issue:43278|issue:43329", - "issue:44977|issue:45563", - "issue:41720|issue:45229", "issue:42490|issue:43575", "issue:10000|issue:15354", + "issue:41720|issue:45229", + "issue:44977|issue:45563", "issue:44568|issue:45491", "issue:45372|issue:45588", "issue:44857|issue:45588", "issue:43412|issue:45588", "issue:45198|issue:45803", "issue:45372|issue:45406", - "issue:43976|issue:45440", - "issue:43278|issue:43381" + "issue:43278|issue:43381", + "issue:43976|issue:45440" ], - "prepared_review_unit_hash": "25f6d3ad6bda0ec7a0320acb04118bc585db9337ae336259dbdc5ca5484ff5e8", + "prepared_review_unit_hash": "4548fdb31031631ecb74d141f908fc81a76b4989bbbf10f1900867dadf49b5d4", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36798, + "serialized_chars": 36797, "estimated_input_tokens": 9200, "estimated_eval_tokens": 18656 }, @@ -8369,192 +8274,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is not a true duplicate cluster: the items span many unrelated issues across tokenization, loading, attention, dtype, OOM, and export paths. All soft duplicate edges look like false positives.", - "confidence": 0.95, - "canonical_issue_reason": "No single canonical issue: the items do not share one underlying defect, so there is no safe umbrella issue to dedupe into.", + "summary": "The cluster is heterogeneous: the soft-linked issues span unrelated tokenizer, model-loading, attention, quantization, export, and runtime errors. None of the soft pairs look like the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the linked issues cover different codepaths and symptoms, so there is no single representative duplicate target.", "canonical_pr_reason": null, - "best_issue_reason": "Issue:45698 is the closest thing to a reusable representative because it concerns from_pretrained/save_pretrained behavior, but it still only covers one serialization path and should not be treated as the cluster canonical.", + "best_issue_reason": "No global best issue: nothing here clearly serves as an umbrella for the cluster, because the soft edges do not indicate a shared root cause.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "Both mention odd model outputs, but they describe different symptoms and likely different code paths; not the same bug." + "reason": "Different problems: one is model-specific zero probabilities, the other is a position-id key warning across many models." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "One is a flash-attention generation issue, the other is a multi-GPU OOM; different failures and fixes." + "reason": "Flash-attention generation failure and multi-GPU CUDA OOM are distinct runtime issues." }, { "left": "issue:44857", "right": "issue:45072", "accept": false, - "reason": "Both are dtype-related, but they affect different models and different inference paths; not one concrete bug." + "reason": "Separate dtype/backend regressions in different models and codepaths; not the same bug." }, { "left": "issue:43746", "right": "issue:46209", "accept": false, - "reason": "PEFT/local checkpoint loading and FP8 inference on B200 are unrelated issues." + "reason": "Local checkpoint loading with PEFT adapters is unrelated to FP8 inference failing on B200." }, { "left": "issue:40444", "right": "issue:43526", "accept": false, - "reason": "IterableDataset multi-image finetuning and BEiT fast processor label reduction are different problems." + "reason": "Multimodal iterable-dataset batching failure is not the same as BEiT processor label reduction." }, { "left": "issue:44448", "right": "issue:44977", "accept": false, - "reason": "Different model families and different regressions: output mismatch vs flash-attention generation failure." + "reason": "Pegasus v4/v5 output drift and Qwen flash-attention generation failure are different defects." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer spacing after added tokens is unrelated to the Trainer _gpu_broadcast NoneType error." + "reason": "Tokenizer spacing after added tokens has nothing to do with Trainer GPU broadcast NoneType errors." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint key mismatch and helper-length behavior are separate issues with no shared failing path." + "reason": "Checkpoint resume key mismatch is unrelated to the qwen3_omni_moe helper-function bug." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Trainer broadcast NoneType and MacOS bus error are different runtime failures." + "reason": "Trainer broadcast NoneType and MacOS bus error on CLIP are different failure modes." }, { - "left": "issue:43550", - "right": "issue:45083", + "left": "issue:29942", + "right": "issue:35141", "accept": false, - "reason": "torch.compile/SDPA failure is not the same as a feature-length helper bug." + "reason": "FlashAttention test failures and embedding reinitialization after resize are unrelated." }, { - "left": "issue:29942", - "right": "issue:35141", + "left": "issue:43550", + "right": "issue:45083", "accept": false, - "reason": "Flash Attention 2 test failures and embedding reinitialization in post_init are unrelated." + "reason": "torch.compile/SDPA failure in Bamba is not the same as undefined helper names in qwen3_omni_moe." }, { "left": "issue:43723", "right": "issue:44554", "accept": false, - "reason": "Tokenizer loading regression and MPS attention correctness are distinct problems." + "reason": "AutoTokenizer v5 loading behavior and an MPS attention correctness issue are different problems." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "Removal of is_torch_fx_available and Deepseek tokenizer output changes do not describe the same bug." + "reason": "Removal of an internal utility breaking trust_remote_code is unrelated to DeepSeek tokenizer output regression." }, { "left": "issue:37428", "right": "issue:44987", "accept": false, - "reason": "Missing flash-support import and loading-physical-intelligence/fast failure are different import/load regressions." + "reason": "Both are loading/import regressions, but they point to different concrete failures and codepaths." }, { - "left": "issue:43873", - "right": "issue:45229", + "left": "issue:43927", + "right": "issue:45698", "accept": false, - "reason": "Quantization offloading behavior and Gemma4 multi-GPU OOM are not the same defect." + "reason": "Save/load state loss of custom token IDs is not the same as loading the wrong custom module after save_pretrained." }, { - "left": "issue:43927", - "right": "issue:45698", + "left": "issue:43873", + "right": "issue:45229", "accept": false, - "reason": "Both involve save/load flows, but one loses custom token IDs while the other loads the wrong custom module; separate bugs." + "reason": "Quantization offloading behavior and a multi-GPU OOM are different memory-management issues." }, { "left": "issue:44462", "right": "issue:44561", "accept": false, - "reason": "AutoTokenizer ignoring tokenizer.json is unrelated to the v5 FX helper removal regression." + "reason": "Tokenizer.json handling and a missing utility function are unrelated." }, { "left": "issue:43278", "right": "issue:43329", "accept": false, - "reason": "Embedding dtype drift in eval and an undefined video-branch helper are unrelated." + "reason": "Embedding dtype drift between train/eval is not the same as undefined video-token helper names." }, { - "left": "issue:44977", - "right": "issue:45563", + "left": "issue:42490", + "right": "issue:43575", "accept": false, - "reason": "Flash-attention generation crash and stale num_return_sequences warning are different generate() issues." + "reason": "Import warning for trailing-slash paths and tensor-parallel OOM are unrelated." }, { - "left": "issue:41720", - "right": "issue:45229", + "left": "issue:10000", + "right": "issue:15354", "accept": false, - "reason": "cudaErrorAssert on auto device mapping is not the same as a multi-GPU OOM." + "reason": "Different historical bugs: German DistilBert issue vs JIT GeneratorExp export failure." }, { - "left": "issue:42490", - "right": "issue:43575", + "left": "issue:41720", + "right": "issue:45229", "accept": false, - "reason": "Trailing-slash import warning and tensor-parallel OOM are unrelated loading issues." + "reason": "Auto device-mapping CUDA assert and multi-GPU OOM are distinct failures." }, { - "left": "issue:10000", - "right": "issue:15354", + "left": "issue:44977", + "right": "issue:45563", "accept": false, - "reason": "German DistilBertModel issue and GeneratorExp torch.jit.export failure are unrelated model/export bugs." + "reason": "Flash-attention generation malfunction and a stale warning for num_return_sequences are not the same issue." }, { "left": "issue:44568", "right": "issue:45491", "accept": false, - "reason": "Tokenizer special-token behavior and Gemma3 NaN embeddings are different bugs." + "reason": "Special-token insertion bug and Gemma3 NaNs under sliding-window attention are different defects." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Gemma4 processor import failure and flash_attention crash on s_aux=None happen in different code paths." + "reason": "Gemma4 processor import failure is unrelated to a flash_attention AttributeError on s_aux=None." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "LwDetr mixed-precision crash and flash_attention sink-less-model crash are unrelated." + "reason": "Loss/AMP crash and flash_attention sink-less-model crash are different codepaths." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Gemma3n executorch export failure and flash_attention AttributeError are different issues." + "reason": "Executorch export failures are unrelated to the flash_attention s_aux=None attribute error." }, { "left": "issue:45198", "right": "issue:45803", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure and Fuyu bare-except behavior are unrelated." + "reason": "Wav2Vec2 save/tokenization failure and a bare-except swallowing errors in Fuyu are distinct bugs." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Both concern Gemma4 processors, but one is an import dependency error and the other is a missing _tokenizer at serve time." + "reason": "Both touch Gemma4, but one is an import dependency issue and the other is a missing _tokenizer attribute in serve." }, { "left": "issue:43976", "right": "issue:45440", "accept": false, - "reason": "Python version compatibility and DeepseekV3MoE divergence are unrelated." - }, - { - "left": "issue:43278", - "right": "issue:43381", - "accept": false, - "reason": "Embedding dtype mismatch in eval and gradient-checkpointing-in-eval-mode are separate problems." + "reason": "Python version compatibility and DeepSeekV3MoE output divergence are unrelated." } ] }, @@ -8629,25 +8528,25 @@ ], "soft_pairs": [ "issue:45588|issue:45823", - "issue:45459|issue:45712", "issue:44448|issue:44466", - "issue:45245|issue:45563", - "issue:43653|issue:44743", "issue:25251|issue:32090", + "issue:45459|issue:45712", + "issue:43653|issue:44743", + "issue:45245|issue:45563", "issue:44206|issue:44479", "issue:43065|issue:43994", "issue:45372|issue:45459", "issue:44186|issue:45198", "issue:43065|issue:45412", - "issue:44871|issue:44977", "issue:43611|issue:46121", + "issue:44871|issue:44977", "issue:32090|issue:33290", "issue:44361|issue:45198", "issue:42490|issue:43873", "issue:33357|issue:39692", - "issue:42915|issue:42994", - "issue:29127|issue:34567", "issue:29942|issue:30064", + "issue:29127|issue:34567", + "issue:42915|issue:42994", "issue:43994|issue:45412", "issue:44265|issue:45198", "issue:43066|issue:44462", @@ -8657,202 +8556,202 @@ "issue:43381|issue:43701", "issue:43653|issue:44442" ], - "prepared_review_unit_hash": "2ab4d983508109d36406734f088f30d2ca5a9a943dfdfba858fd542aea0d2350", + "prepared_review_unit_hash": "f79b8f677af473d96df026ffa27938c17ca92c5437efe37e9878e0142f459d1b", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36174, - "estimated_input_tokens": 9044, - "estimated_eval_tokens": 18344 + "serialized_chars": 36178, + "estimated_input_tokens": 9045, + "estimated_eval_tokens": 18346 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft pair looks like a true duplicate; the candidates share broad subsystems at most, but each points to a different bug or regression.", - "confidence": 0.96, - "canonical_issue_reason": "No clear canonical issue emerges because the set is heterogeneous and the soft pairs are about distinct problems, not one underlying defect.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports; the only loose common thread is tokenizer/loading regressions, and none of the soft pairs look like true duplicates. I\u2019d use 44462 as the closest canonical issue because it is the broadest user-facing loading regression in that subset.", + "confidence": 0.32, + "canonical_issue_reason": "44462 is the most broadly scoped, user-facing tokenizer-loading regression here, so it best represents the small cluster of tokenizer-related reports better than the more niche failures.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a safe umbrella representative here; choosing one would be arbitrary since none of the listed bugs subsumes the others.", + "best_issue_reason": "44462 is the best representative issue overall because it captures a common loading-path failure with wider duplicate potential than the more model-specific or one-off bugs in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "Different failures in Gemma-family code: one is flash-attention handling of `s_aux=None`, the other is a device-map forward mismatch." + "reason": "Different code paths and symptoms: flash-attention handling of s_aux=None versus a Gemma4 device_map/device-placement mismatch." }, { - "left": "issue:45459", - "right": "issue:45712", + "left": "issue:44448", + "right": "issue:44466", "accept": false, - "reason": "Tokenizer error masking from missing protobuf is unrelated to leftover dummy PT classes and repo-check leakage." + "reason": "Both are v5 regressions, but one is Pegasus generation output drift and the other is serialization of lm_head.weight; not the same concrete bug." }, { - "left": "issue:44448", - "right": "issue:44466", + "left": "issue:25251", + "right": "issue:32090", "accept": false, - "reason": "One is a model output difference across v4/v5, the other is inconsistent serialization of tied weights depending on device." + "reason": "Pipeline top_k nesting and Trainer _gpu_broadcast_one NoneType errors are unrelated failures." }, { - "left": "issue:45245", - "right": "issue:45563", + "left": "issue:45459", + "right": "issue:45712", "accept": false, - "reason": "A category-count runtime error is not the same as a stale warning in paged `generate()`." + "reason": "Tokenizer exception masking when protobuf is missing is unrelated to cleanup of leftover dummy classes." }, { "left": "issue:43653", "right": "issue:44743", "accept": false, - "reason": "BigBirdTokenizer special-token registration and Qwen3.5 recurrent-state reset are unrelated code paths." + "reason": "BigBirdTokenizer special-token decode bug and Qwen recurrent-state reset logic are different subsystems and failure modes." }, { - "left": "issue:25251", - "right": "issue:32090", + "left": "issue:45245", + "right": "issue:45563", "accept": false, - "reason": "Pipeline `top_k` nesting behavior is unrelated to Trainer broadcast failures on `NoneType`." + "reason": "Quantized-save category limit and a stale generate() warning for num_return_sequences are not the same bug." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "A feature-extractor `center` argument crash is not the same as the Qwen video-input regression." + "reason": "Unsupported center arg in a feature extractor is distinct from the Qwen video-input regression." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Dummy `Conv2d` in Sam3PixelDecoder is unrelated to SigLIP2 nonsense outputs from AutoModel/pipeline loading." + "reason": "Sam3PixelDecoder dummy Conv2d and SigLIP2 nonsensical outputs are different concrete model issues." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma 4 processor import failure due to `ReasoningEffort` is unrelated to protobuf-masked tokenizer errors." + "reason": "Gemma4 processor import failure due to ReasoningEffort is unrelated to protobuf-related tokenizer error masking." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2 tokenizer crashes on NER/padding are different from Wav2Vec2 save/tokenization failures." + "reason": "Both involve tokenizers, but LayoutLMv2 NER/padding crashes and Wav2Vec2 save_pretrained/tokenization failures are different bugs." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "A dummy layer definition is not the same issue as RT-DETR models not releasing memory on deletion." + "reason": "A dummy layer in Sam3PixelDecoder is not the same as RT-DETR memory not being released after deletion." }, { - "left": "issue:44871", - "right": "issue:44977", + "left": "issue:43611", + "right": "issue:46121", "accept": false, - "reason": "An EOS config mismatch in Gemma-3 is unrelated to Qwen3.5 flash-attention generation failures." + "reason": "base_model_prefix loading regression and JSON-loaded rope validation list TypeError affect different code paths." }, { - "left": "issue:43611", - "right": "issue:46121", + "left": "issue:44871", + "right": "issue:44977", "accept": false, - "reason": "Base-model-prefix loading breakage is a different bug from `convert_rope_params_to_dict` rejecting JSON-loaded lists." + "reason": "Gemma-3 eos_token_id config mismatch and Qwen3.5 flash-attention generation failure are unrelated." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Trainer broadcast of `NoneType` and AdaFactor+DeepSpeed OOM are distinct runtime problems." + "reason": "NoneType broadcast failure in Trainer is not the same as an OOM with Adafactor in DeepSpeed." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "MLukeTokenizer task-time AttributeError is unrelated to Wav2Vec2 save/tokenization failure." + "reason": "MLukeTokenizer AttributeError and Wav2Vec2 save/tokenization failure are different tokenizer bugs." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "A trailing-slash local-path warning is not the same as quantization offloading misbehavior." + "reason": "Local-path trailing-slash import warning and quantization offloading failure are not the same issue." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "A MacOS bus error on a community CLIP model is different from SigLIP2 documentation example errors." + "reason": "A MacOS bus error on a CLIP model and SigLIP2 documentation example errors are different problems." }, { - "left": "issue:42915", - "right": "issue:42994", + "left": "issue:29942", + "right": "issue:30064", "accept": false, - "reason": "FineGrainedFP8Config failure for Qwen3Moe and quantized-model saving failure are different bugs." + "reason": "Flash Attention 2 test failures and void segmentation-map processing failures do not share the same root cause." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "LayoutLMv3 error-message clarity and `num_input_tokens_seen` not updating are unrelated Trainer/layout issues." + "reason": "LayoutLMv3 error clarity and TrainerState num_input_tokens_seen not updating are unrelated." }, { - "left": "issue:29942", - "right": "issue:30064", + "left": "issue:42915", + "right": "issue:42994", "accept": false, - "reason": "Failing Flash Attention 2 tests are not the same as image-processor void-segmentation-map handling." + "reason": "FineGrainedFP8Config failure and quantized-model saving failure are distinct quantization bugs." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "SigLIP2 wrong outputs and RT-DETR memory retention do not share the same underlying code-path bug." + "reason": "SigLIP2 wrong outputs and RT-DETR memory leaks are different symptoms and code paths." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "`torch.export.export` incompatibility with `torch_compilable_check` is unrelated to Wav2Vec2 serialization/tokenization issues." + "reason": "torch.export.export failure with torch_compilable_check is unrelated to Wav2Vec2 tokenization/save_pretrained failures." }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Wrong tokenizer decoder type and ignoring `tokenizer.json` are related to tokenizer loading broadly, but they are not the same concrete defect." + "reason": "Both are tokenizer-loading regressions, but one is a decoder-type mismatch while the other is AutoTokenizer ignoring tokenizer.json; not the same concrete bug." }, { "left": "issue:43257", "right": "issue:43278", "accept": false, - "reason": "MOE weight conversion under accelerate+deepspeed is unrelated to embedding dtype changing between train and eval." + "reason": "Missing MOE weight conversion under accelerate+deepspeed and BF16-to-FP32 dtype drift are different issues." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Smollm3 RoPE-layer omission and DeepseekV3MoE divergence are different model-specific implementation mismatches." + "reason": "Smollm3 dropping RoPE layers and DeepseekV3MoE diverging from remote implementation are architecture-specific and not the same bug." }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "Runtime special-token post-processing updates are unrelated to Gemma3 NaN embeddings with sliding-window attention." + "reason": "Runtime post-processor updates for special tokens and Gemma3 NaN embeddings with mixed-length batching are unrelated." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Gradient checkpointing being disallowed in eval mode is a different issue from resume-from-checkpoint key mismatches." + "reason": "Eval-mode gradient checkpointing restriction and resume_from_checkpoint key mismatch are separate training-state problems." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "BigBirdTokenizer special-token registration and AutoTokenizer loading FastSpeech2ConformerTokenizer are separate tokenizer bugs." + "reason": "Both are tokenizer-related, but one is special-token registration/empty decode and the other is AutoTokenizer class loading; different root causes." } ] }, @@ -8886,7 +8785,7 @@ "issue:43262", "issue:43334", "issue:43381", - "issue:43441", + "issue:43452", "issue:43526", "issue:43531", "issue:43575", @@ -8910,7 +8809,6 @@ "issue:44811", "issue:44857", "issue:44977", - "issue:45003", "issue:45072", "issue:45103", "issue:45245", @@ -8928,222 +8826,215 @@ "soft_pairs": [ "issue:44610|issue:45412", "issue:43824|issue:44279", - "issue:45440|issue:45593", "issue:29127|issue:33357", + "issue:45440|issue:45593", "issue:29942|issue:36010", - "issue:43526|issue:45834", "issue:43761|issue:45859", + "issue:43526|issue:45834", "issue:43526|issue:44186", "issue:45103|issue:45372", - "issue:44625|issue:45584", "issue:29127|issue:35141", - "issue:43334|issue:43531", + "issue:44625|issue:45584", "issue:44442|issue:45072", + "issue:43334|issue:43531", "issue:42915|issue:43931", - "issue:44977|issue:45245", + "issue:43575|issue:43701", "issue:43526|issue:44857", "issue:42490|issue:43381", - "issue:43575|issue:43701", + "issue:44977|issue:45245", "issue:43575|issue:44387", "issue:43262|issue:44206", "issue:44464|issue:45491", "issue:43646|issue:44077", "issue:44743|issue:44811", - "issue:43441|issue:45588", + "issue:43452|issue:45588", "issue:44062|issue:45907", - "issue:43761|issue:44079", - "issue:45003|issue:45440" + "issue:43761|issue:44079" ], - "prepared_review_unit_hash": "4ef9029a71e088c838ea7ca84c413b54718b8284140b45e8114cb1df82761cdb", + "prepared_review_unit_hash": "6629c151882e98d7a8fb5e428079005988cc3918d8f3bb7b52be1c8a01d22c2c", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 27, - "serialized_chars": 35731, - "estimated_input_tokens": 8933, - "estimated_eval_tokens": 18122 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34908, + "estimated_input_tokens": 8727, + "estimated_eval_tokens": 17710 }, "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. A few pairs are thematically adjacent, but none are convincing duplicates from the available titles alone.", - "confidence": 0.68, - "canonical_issue_reason": "No true duplicate anchor stands out; if one must be chosen, issue #43646 is the broadest framework-level regression and the best representative of the cluster\u2019s few initialization/regression themes.", + "summary": "The soft pairs are mostly unrelated issues across different models and subsystems; none look like the same underlying bug or change, so no pair should be merged as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "No single issue is a clear duplicate anchor for this set: the reports span unrelated regressions and components. If a representative must be chosen, issue 43646 is the broadest, but it is not a true canonical duplicate target.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #43646 is the most generally useful anchor: it describes a broad Transformers 5.0.0 regression affecting custom model initialization, whereas the others are narrower model- or component-specific bugs.", + "best_issue_reason": "Issue 43646 is the most general, high-level regression in the set (a broad Transformers 5.0.0 initialization break), so it is the best representative bucket if one is needed, though the cluster is not actually homogeneous.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44610", "right": "issue:45412", "accept": false, - "reason": "Different problems: processor size mismatch vs memory leak on deletion." + "reason": "Different failures: processor input size mismatch for OmDet-Turbo vs RT-DETR memory not being released." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "Both mention imports/dependencies, but they describe different failure modes and fixes." + "reason": "One is a specific missing import for Qwen2.5-VL; the other is a broad dependency issue with no concrete same bug." }, { - "left": "issue:45440", - "right": "issue:45593", + "left": "issue:29127", + "right": "issue:33357", "accept": false, - "reason": "Unrelated model behavior bugs in different components." + "reason": "LayoutLMv3 error messaging clarity and a MacOS bus error on CLIP are unrelated." }, { - "left": "issue:29127", - "right": "issue:33357", + "left": "issue:45440", + "right": "issue:45593", "accept": false, - "reason": "Different models and completely different failure types." + "reason": "DeepSeekV3MoE implementation divergence and D-FINE auxiliary-loss behavior are different code paths." }, { "left": "issue:29942", "right": "issue:36010", "accept": false, - "reason": "Flash Attention test failures are unrelated to the GenerationMixin import error." + "reason": "Flash Attention 2 test failures are not the same issue as a GenerationMixin import error." }, { - "left": "issue:43526", - "right": "issue:45834", + "left": "issue:43761", + "right": "issue:45859", "accept": false, - "reason": "Both are OCR/vision-related, but the label-reduction bug is not the same as a long-input index error." + "reason": "CLIPVision hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are separate bugs." }, { - "left": "issue:43761", - "right": "issue:45859", + "left": "issue:43526", + "right": "issue:45834", "accept": false, - "reason": "Different model families and unrelated runtime issues." + "reason": "BEiT FastImageProcessor label reduction and Kosmos2.5 long OCR index errors are unrelated." }, { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Both involve tokenizers/processors, but they are distinct bugs with different models and code paths." + "reason": "These affect different tokenizer/processor pipelines and different failure modes." }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "Docstring/introspection crash vs an external import breakage." - }, - { - "left": "issue:44625", - "right": "issue:45584", - "accept": false, - "reason": "Different model-specific regressions with no shared code path." + "reason": "auto_docstring annotation handling and Gemma 4 loading/import issues are not the same bug." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "Error-message clarity issue versus embedding reinitialization bug." + "reason": "LayoutLMv3 box validation messaging and token-embedding reinitialization after resize are unrelated." }, { - "left": "issue:43334", - "right": "issue:43531", + "left": "issue:44625", + "right": "issue:45584", "accept": false, - "reason": "Same broad Qwen family, but different subsystems and failure mechanisms." + "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription generation failure are different problems." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Tokenizer loading and dtype mismatch are unrelated." + "reason": "FastSpeech2Conformer tokenizer loading and bfloat16 dtype mismatches are distinct issues." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id for Qwen3-VL loading is not the same as Qwen3-MoE sliding-window behavior." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "Different Qwen model/config failures, not the same bug." + "reason": "FineGrainedFP8Config failure for Qwen3Moe and Qwen3-VL weight-shape mismatch are different model/config bugs." }, { - "left": "issue:44977", - "right": "issue:45245", + "left": "issue:43575", + "right": "issue:43701", "accept": false, - "reason": "Flash-attention generation bug and category-cardinality error are unrelated." + "reason": "Tensor-parallel OOM on Qwen2-57B is unrelated to resume_from_checkpoint key mismatch." }, { "left": "issue:43526", "right": "issue:44857", "accept": false, - "reason": "Both are vision/detection bugs, but they are separate failure modes." + "reason": "Label reduction bug in BeitImageProcessorFast and float16 AMP crash in LwDetrImageLoss are separate." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Local-path import warning and eval-mode gradient checkpointing are unrelated." + "reason": "Trailing-slash local model loading warning is unrelated to gradient checkpointing being disallowed in eval mode." }, { - "left": "issue:43575", - "right": "issue:43701", + "left": "issue:44977", + "right": "issue:45245", "accept": false, - "reason": "Model loading OOM and checkpoint key mismatch are unrelated." + "reason": "Flash-attention generation problems and the 2^24 category limit are unrelated." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Both concern memory, but the concrete causes and code paths differ." + "reason": "Both mention OOM, but one is tensor-parallel model loading and the other is int4 quantization memory growth; different root causes." }, { "left": "issue:43262", "right": "issue:44206", "accept": false, - "reason": "Both are audio preprocessing issues, but they affect different APIs and different bugs." + "reason": "Audio chat-template sampling-rate defaults and an unsupported center argument in LasrFeatureExtractor are unrelated." }, { "left": "issue:44464", "right": "issue:45491", "accept": false, - "reason": "Generation correctness issues, but on different models and different mechanisms." + "reason": "Compiled-forward chunked generation inconsistency and Gemma3 NaN embeddings from sliding-window padding are different bugs." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "Related to model initialization/post_init, but not clearly the same underlying bug or fix." + "reason": "Broad custom model initialization regression and a patchtsmixer post_init policy issue do not share the same concrete failure." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Qwen recurrent-state bug and Whisper decode option handling are unrelated." + "reason": "Qwen3.5 recurrent-state caching bug and Whisper batch_decode skip_special_tokens handling are unrelated." }, { - "left": "issue:43441", + "left": "issue:43452", "right": "issue:45588", "accept": false, - "reason": "Both touch FlashAttention, but they involve different components and distinct crashes." + "reason": "gguf_file loading behavior and flash_attention sink-less AttributeError are separate code paths." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "Tokenizer AddedToken error and list_repo_files revision forwarding are unrelated." + "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are unrelated API bugs." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "The ModelOutput title is superficially relevant, but the CLIPVision hidden_states regression is not clearly the same bug." - }, - { - "left": "issue:45003", - "right": "issue:45440", - "accept": false, - "reason": "Unsafe sys.modules access and DeepSeek divergence are unrelated." + "reason": "CLIPVision hidden_states regression and ModelOutput key assignment with None are different issues." } ] }, @@ -9167,8 +9058,6 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ - "issue:25251", - "issue:28282", "issue:33290", "issue:34567", "issue:34689", @@ -9205,9 +9094,11 @@ "issue:44898", "issue:44913", "issue:44987", + "issue:45003", "issue:45137", "issue:45216", "issue:45290", + "issue:45440", "issue:45479", "issue:45698", "issue:45759", @@ -9217,245 +9108,245 @@ "issue:46209" ], "soft_pairs": [ - "issue:43723|issue:45479", "issue:33290|issue:34689", "issue:44442|issue:44898", + "issue:45003|issue:45440", "issue:33290|issue:34567", - "issue:43232|issue:43388", "issue:33290|issue:35141", + "issue:43723|issue:45479", + "issue:43232|issue:43388", + "issue:43381|issue:43873", "issue:42490|issue:44743", "issue:43316|issue:43525", - "issue:43381|issue:43873", "issue:43761|issue:44188", - "issue:44479|issue:45290", "issue:34689|issue:43611", - "issue:44987|issue:46178", + "issue:44479|issue:45290", "issue:43720|issue:44442", - "issue:44442|issue:45803", + "issue:44987|issue:46178", "issue:44077|issue:44479", - "issue:43931|issue:45216", + "issue:44442|issue:45803", "issue:43493|issue:43827", + "issue:43931|issue:45216", "issue:34689|issue:36010", "issue:42907|issue:45698", "issue:33290|issue:36331", "issue:43316|issue:45759", "issue:45290|issue:46209", - "issue:43749|issue:43931", "issue:43873|issue:44871", + "issue:43749|issue:43931", "issue:43643|issue:44913", "issue:43650|issue:46202", "issue:44190|issue:44496", - "issue:44855|issue:45137", - "issue:25251|issue:28282" + "issue:44855|issue:45137" ], - "prepared_review_unit_hash": "e6ca3bd2dd4dc8669fc9e3b98f7323f82eeac8c6e298612455d6c82a24a3c635", + "prepared_review_unit_hash": "7385df6bf9cca8b91f177942b6cdbce531036e9f2138c63cc06b7bcdc290b26c", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36936, - "estimated_input_tokens": 9234, - "estimated_eval_tokens": 18724 + "serialized_chars": 36890, + "estimated_input_tokens": 9223, + "estimated_eval_tokens": 18702 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is heterogeneous and does not form a duplicate cluster; all soft-edge pairs are different bugs/changes, often in different subsystems or with different failure modes.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue stands out: these issues cover unrelated regressions across loading, generation, config handling, training, and docs.", + "summary": "The set is highly heterogeneous and the soft edges are mostly superficial title-level similarities rather than true duplicates. I would reject every proposed merge pair.", + "confidence": 0.83, + "canonical_issue_reason": "No single canonical issue stands out: these reports span unrelated bugs across loading, configs, training, generation, vision, quantization, and docs, so there is no clear duplicate anchor for the whole set.", "canonical_pr_reason": null, - "best_issue_reason": "No clear best anchor issue for deduping the whole set; picking any one would conflate distinct bug classes.", + "best_issue_reason": "Issue 43873 is the strongest standalone anchor among the set because it is currently open, detailed, and has the most discussion/inbound references, but it still is not related to most of the other items.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43723", - "right": "issue:45479", - "accept": false, - "reason": "Tokenizer-loading regression and zero-loss classification bug are unrelated code paths." - }, { "left": "issue:33290", "right": "issue:34689", "accept": false, - "reason": "DeepSpeed Adafactor OOM and Llama 3.2 vision model-loading failure are different issues." + "reason": "Different bugs: DeepSpeed/Adafactor OOM vs Llama 3.2 Vision model loading regression." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "Tokenizer import/load failure and Perceiver positional-encoding failure are unrelated." + "reason": "Unrelated failures: tokenizer loading vs Perceiver image classification resize/interpolation bug." }, { - "left": "issue:33290", - "right": "issue:34567", + "left": "issue:45003", + "right": "issue:45440", "accept": false, - "reason": "Optimizer OOM and TrainerState token counter not updating are distinct training bugs." + "reason": "Different code paths: unsafe sys.modules access vs DeepSeekV3MoE behavioral divergence." }, { - "left": "issue:43232", - "right": "issue:43388", + "left": "issue:33290", + "right": "issue:34567", "accept": false, - "reason": "Generation kwargs syncing and metric-gathering label truncation affect different paths." + "reason": "OOM in optimizer/deepspeed is unrelated to TrainerState token counter not updating." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "Deepspeed optimizer OOM is unrelated to embedding reinitialization after resize." + "reason": "Optimizer memory blowup is unrelated to embedding resize/post_init reinitialization." }, { - "left": "issue:42490", - "right": "issue:44743", + "left": "issue:43723", + "right": "issue:45479", "accept": false, - "reason": "Local-model path warning and recurrent-state reset in Qwen3.5 are different bugs." + "reason": "Tokenizer loading regression and degenerate classification loss are different problems." }, { - "left": "issue:43316", - "right": "issue:43525", + "left": "issue:43232", + "right": "issue:43388", "accept": false, - "reason": "Both mention config API issues, but Gemma3TextConfig and Llama4Config failures are not the same bug." + "reason": "Generation sync/update bug is unrelated to metric gathering label truncation." }, { "left": "issue:43381", "right": "issue:43873", "accept": false, - "reason": "Eval-mode gradient checkpointing and quantized offloading failures are unrelated." + "reason": "Gradient checkpointing eval-mode restriction is unrelated to quantization/offloading behavior." }, { - "left": "issue:43761", - "right": "issue:44188", + "left": "issue:42490", + "right": "issue:44743", "accept": false, - "reason": "CLIP hidden_states regression and attention-kernel divergence under torch.compile are different code paths." + "reason": "Local-path import warning and recurrent-state cache reset are unrelated." }, { - "left": "issue:44479", - "right": "issue:45290", + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both mention config API issues, but different models and different missing fields." + }, + { + "left": "issue:43761", + "right": "issue:44188", "accept": false, - "reason": "Video-input regression and chat-template tool-call crash are separate multimodal bugs." + "reason": "CLIP hidden_states regression is unrelated to attention-kernel divergence under torch.compile." }, { "left": "issue:34689", "right": "issue:43611", "accept": false, - "reason": "Both are model-loading regressions, but they target different mechanisms and models." + "reason": "Both are model-loading regressions, but they concern different mechanisms and different failure modes." }, { - "left": "issue:44987", - "right": "issue:46178", + "left": "issue:44479", + "right": "issue:45290", "accept": false, - "reason": "Physical-intelligence loading regression and tensorflow_text backend mapping error are unrelated." + "reason": "Video-input regression and chat-template tool-call crash are separate bugs." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight loading and FastSpeech2 tokenizer loading are different failures." + "reason": "BitNet accelerate loading issue is unrelated to tokenizer loading failure." }, { - "left": "issue:44442", - "right": "issue:45803", + "left": "issue:44987", + "right": "issue:46178", "accept": false, - "reason": "Tokenizer loading and Fuyu tensor-conversion exception handling are unrelated." + "reason": "Different symptoms and subsystems: model loading regression vs missing BACKENDS_MAPPING entry." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "PatchTSMixer post_init policy and Qwen video regression do not share the same bug." + "reason": "Optional post_init contract issue is unrelated to video input regression." }, { - "left": "issue:43931", - "right": "issue:45216", + "left": "issue:44442", + "right": "issue:45803", "accept": false, - "reason": "Weight-shape mismatch on load and incorrect checkpoint save behavior are not the same defect." + "reason": "Tokenizer load failure and bare-except error swallowing in Fuyu are unrelated." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 implementation mismatch and docs still referencing pipeline() are unrelated." + "reason": "Model implementation discrepancy and docs still referencing pipeline() are not the same bug." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Shape-mismatch load error and save_pretrained checkpoint corruption are distinct issues." }, { "left": "issue:34689", "right": "issue:36010", "accept": false, - "reason": "Model loading regression and GenerationMixin import error are different import/load failures." + "reason": "Model loading breakage and GenerationMixin import error are different failures." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Dequantized save failure and wrong custom-module reload are distinct save/load issues." + "reason": "Dequantized save failure and wrong custom module reloading are separate save/load problems." }, { "left": "issue:33290", "right": "issue:36331", "accept": false, - "reason": "OOM in optimizer setup and compute_loss signature breakage are unrelated training problems." + "reason": "DeepSpeed OOM and CustomTrainer compute_loss signature mismatch are unrelated training bugs." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Gemma3TextConfig API discrepancy and Qwen text_config unwrapping bug are different config issues." + "reason": "Config API discrepancy and AutoModelForCausalLM unwrapping behavior are different issues." }, { "left": "issue:45290", "right": "issue:46209", "accept": false, - "reason": "Chat template crash and FP8 inference failure on B200 are unrelated." + "reason": "Chat template crash is unrelated to FP8 inference failure on B200." }, { - "left": "issue:43749", - "right": "issue:43931", + "left": "issue:43873", + "right": "issue:44871", "accept": false, - "reason": "FSDP CPU-efficient loading and Qwen3-VL shape mismatch are different loading regressions." + "reason": "Quantization/offloading bug and Gemma eos_token_id inconsistency are unrelated." }, { - "left": "issue:43873", - "right": "issue:44871", + "left": "issue:43749", + "right": "issue:43931", "accept": false, - "reason": "Quantized offloading bug and Gemma eos_token_id inconsistency are unrelated config/runtime issues." + "reason": "FSDP CPU RAM efficient loading regression is not the same as Qwen3-VL shape mismatch." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "Remote-code AutoConfig field loss and GPTNeoX rotary_pct reload default are similar in theme but not the same bug." + "reason": "trust_remote_code missing-fields issue and rotary_pct reload regression are different config serialization bugs." }, { "left": "issue:43650", "right": "issue:46202", "accept": false, - "reason": "Placeholder issue and Gemma pipeline roundtrip failure are unrelated." + "reason": "Placeholder/noise issue is unrelated to Gemma pipeline roundtrip failure." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading script bug and unrecognized model config error are different failures." + "reason": "Local dataset loading problem is unrelated to unrecognized custom model config." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 import parsing error and ZeRO3 deque IndexError are unrelated." - }, - { - "left": "issue:25251", - "right": "issue:28282", - "accept": false, - "reason": "Pipeline output nesting bug and missing-PyTorch import error are clearly unrelated." + "reason": "Python 3.13 import parsing error and DeepSpeed ZeRO3 deque pop error are unrelated." } ] }, @@ -9481,6 +9372,8 @@ "nodes": [ "issue:16998", "issue:22355", + "issue:25251", + "issue:28282", "issue:32090", "issue:35141", "issue:41950", @@ -9493,7 +9386,6 @@ "issue:43441", "issue:43452", "issue:43475", - "issue:43504", "issue:43540", "issue:43550", "issue:43572", @@ -9522,31 +9414,31 @@ "issue:45127", "issue:45200", "issue:45584", - "issue:45698", "issue:45803", "issue:46050", "issue:46170", "issue:46178" ], "soft_pairs": [ + "issue:25251|issue:28282", "issue:43295|issue:43825", "issue:43065|issue:44617", "issue:43550|issue:43701", + "issue:43606|issue:43701", "issue:43316|issue:44877", "issue:43316|issue:43716", - "issue:43606|issue:43701", + "issue:42617|issue:43575", "issue:43540|issue:44514", "issue:44077|issue:44534", - "issue:42617|issue:43575", "issue:44162|issue:44792", "issue:16998|issue:22355", "issue:44871|issue:45584", + "issue:43575|issue:43742", "issue:43404|issue:43441", "issue:45803|issue:46170", - "issue:43575|issue:43742", + "issue:43295|issue:44871", "issue:43673|issue:44743", "issue:41950|issue:43475", - "issue:43295|issue:44871", "issue:43761|issue:44479", "issue:43388|issue:43688", "issue:43452|issue:44496", @@ -9554,199 +9446,198 @@ "issue:44871|issue:46050", "issue:45127|issue:45200", "issue:43572|issue:46178", - "issue:32090|issue:35141", - "issue:43504|issue:45698" + "issue:32090|issue:35141" ], - "prepared_review_unit_hash": "06f9b43e2d84ec8620798e45bd38fbac774daacaa2cc4b8e711fc94ee5d0651b", + "prepared_review_unit_hash": "9cb918df43bbb13c11e4be14696883a850a55d8a56b194d35689caadb0a03657", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 27, - "serialized_chars": 35716, - "estimated_input_tokens": 8929, - "estimated_eval_tokens": 18114 + "serialized_chars": 35722, + "estimated_input_tokens": 8931, + "estimated_eval_tokens": 18118 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a grab-bag of unrelated Transformers bug reports rather than a duplicate set. None of the soft edges look like the same underlying bug/change, so all should be rejected.", - "confidence": 0.9, - "canonical_issue_reason": "There is no true canonical duplicate in this mixed cluster; if one representative must be chosen, issue 45698 is the broadest and most active open report around save/load behavior after save_pretrained.", + "summary": "This cluster is a loose collection of unrelated bug reports across many subsystems; the soft-similarity pairs do not look like true duplicates or mergeable same-fix candidates. No PRs are present.", + "confidence": 0.96, + "canonical_issue_reason": "issue:43646 is the most general and high-level regression in the set (custom model initialization breaking under Transformers 5.0), so it is the closest thing to a cluster anchor, though the cluster is not a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45698 is the best stand-alone representative here because it is open, well-discussed, and concerns a general from_pretrained/save_pretrained regression affecting custom modules.", + "best_issue_reason": "issue:43646 is the best representative issue because it describes a broad, user-facing regression with clear actionability and relatively generic impact compared with the other highly specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Different failures: pipeline output shape change vs missing `transformers.onnx` module import." + }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "Both are v5 regressions, but one is about processor.tokenizer/image handling and the other is a pipeline translation error message. Different code paths and different failures." + "reason": "One is a processor API regression; the other is an incorrect pipeline error message about translation support." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d and Sam3Video CUDA OOM are unrelated symptoms in different SAM 3 components." + "reason": "`Sam3PixelDecoder` dummy layer issue is unrelated to SAM3 video CUDA OOM." }, { "left": "issue:43550", "right": "issue:43701", "accept": false, - "reason": "torch.compile + SDPA failure in Bamba is unrelated to a resume_from_checkpoint key mismatch." + "reason": "torch.compile/SDPA failure in Bamba is unrelated to checkpoint resume key mismatches." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch does not match resume-from-checkpoint key mismatch." }, { "left": "issue:43316", "right": "issue:44877", "accept": false, - "reason": "One is an API discrepancy in Gemma3TextConfig; the other is a strict config loading issue for granite_speech. Different underlying bugs." + "reason": "Gemma3 config API discrepancy is unrelated to strict config loading for granite_speech." }, { "left": "issue:43316", "right": "issue:43716", "accept": false, - "reason": "Gemma3TextConfig API mismatch and Mistral-3 image preprocessor dtype mismatch are separate model-specific issues." + "reason": "Config API mismatch is not the same as an image preprocessor dtype mismatch." }, { - "left": "issue:43606", - "right": "issue:43701", + "left": "issue:42617", + "right": "issue:43575", "accept": false, - "reason": "CPU offload device mismatch for bark-small is unrelated to checkpoint resume key handling." + "reason": "Failed script execution and TP OOM are different problems and code paths." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Both involve Qwen multimodal processing, but one is a video-input ValueError in Qwen3OmniMoe and the other is a batched apply_chat_template crash in Qwen2_5_VLProcessor." + "reason": "Qwen3OmniMoe video-processing error is separate from batched chat-template padding crash." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "patchtsmixer post_init validation and non-persistent buffer junk initialization are different regression classes." - }, - { - "left": "issue:42617", - "right": "issue:43575", - "accept": false, - "reason": "Failure to run 3d_parallel.py and OOM loading Qwen2-57B with tensor parallelism are unrelated." + "reason": "Optional `post_init` policy issue is unrelated to non-persistent buffer initialization bugs." }, { "left": "issue:44162", "right": "issue:44792", "accept": false, - "reason": "ESM2 breakage and a janus image-generation test failure do not share the same underlying bug." + "reason": "ESM2 regression and Janus image-generation test failure do not describe the same underlying bug." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "DeBERTa-V3 model_max_length confusion is unrelated to a missing transformers.onnx module import." + "reason": "Model max length question is unrelated to missing ONNX submodule import." }, { "left": "issue:44871", "right": "issue:45584", "accept": false, - "reason": "Gemma-3 eos_token_id inconsistency and Whisper empty-transcription generation after align_special_tokens are separate issues." + "reason": "Gemma EOS config inconsistency is unrelated to Whisper empty-transcription generation failure." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Tensor-parallel OOM and MobileLLM config key error are different loading failures." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "Mistral3 lm_head weight tying and Ministral-3 FlashAttention failure are different concrete code-path problems." + "reason": "Untied `lm_head` weights and FlashAttention failure in Ministral-3 are different code-path problems." }, { "left": "issue:45803", "right": "issue:46170", "accept": false, - "reason": "A bare except in Fuyu tensor conversion is unrelated to Windows/ROCm import errors around torch.distributed." + "reason": "Bare `except` in Fuyu tensor conversion is unrelated to Windows/ROCm distributed import issues." }, { - "left": "issue:43575", - "right": "issue:43742", + "left": "issue:43295", + "right": "issue:44871", "accept": false, - "reason": "Tensor-parallel OOM on Qwen2-57B and a key error loading MobileLLM-125M are not the same bug." + "reason": "Processor/tokenizer regression is not the same as Gemma EOS token configuration mismatch." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both mention cache/state behavior, but one is a missing GenerationMixin cache in chunked_prefill and the other is recurrent state reset in modular_qwen3_5; different mechanisms and fixes." + "reason": "Cache missing in generation and recurrent state reset are related thematically but not the same concrete bug." }, { "left": "issue:41950", "right": "issue:43475", "accept": false, - "reason": "Video-classification pipeline processor lookup and Sam3Video missing fpn_position_embeddings are unrelated." - }, - { - "left": "issue:43295", - "right": "issue:44871", - "accept": false, - "reason": "processor.tokenizer regression and Gemma-3 eos_token_id config inconsistency are unrelated issues." + "reason": "Video-classification processor selection bug is unrelated to missing `fpn_position_embeddings` on SAM3 output." }, { "left": "issue:43761", "right": "issue:44479", "accept": false, - "reason": "CLIPVisionModel hidden_states regression and Qwen video-input regression affect different models and code paths." + "reason": "CLIP vision hidden-states regression is separate from Qwen video-input regression." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Metric label dropping in last batch and auxiliary loss normalization are both training-related, but they are distinct bugs." + "reason": "Metric label truncation and auxiliary-loss normalization are different training bugs." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "gguf_file loading breakage and unrecognized Olmo-Hybrid config are separate model-loading failures." + "reason": "GGUF loading breakage is different from missing `model_type` in config loading." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "Transformers 5.0 custom model init breakage and GPTNeoX rotary_pct reload behavior are not the same underlying problem." + "reason": "Custom model init regression is broader and different from GPTNeoX rotary_pct persistence on reload." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Gemma-3 eos_token_id mismatch and sinq quantized save/load breakage are unrelated persistence/config issues." + "reason": "Gemma EOS mismatch is unrelated to sinq quantization save/load breakage." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "LoRA merge collapse with tied embeddings and Gemma 4 mm_token_type_ids defaults are different model-training issues." + "reason": "LoRA merge collapse with extended vocab is unrelated to Gemma 4 multimodal token type defaults." }, { "left": "issue:43572", "right": "issue:46178", "accept": false, - "reason": "StableLmConfig pad_token_idx regression and tensorflow_text backend lookup failure are unrelated." + "reason": "StableLm missing config field and tensorflow_text backend mapping error are unrelated compatibility issues." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Trainer GPU broadcast NoneType error and token-embedding resize reinitialization are unrelated bugs." - }, - { - "left": "issue:43504", - "right": "issue:45698", - "accept": false, - "reason": "Legacy-field loading for Beit semantic segmentation and wrong custom module loading after save_pretrained are separate save/load issues with different causes." + "reason": "Trainer GPU broadcast NoneType error is unrelated to embedding reinitialization after resizing." } ] }, @@ -9775,6 +9666,7 @@ "issue:42490", "issue:43065", "issue:43335", + "issue:43504", "issue:43525", "issue:43531", "issue:43540", @@ -9788,7 +9680,6 @@ "issue:44060", "issue:44077", "issue:44079", - "issue:44112", "issue:44190", "issue:44242", "issue:44464", @@ -9815,249 +9706,243 @@ "issue:45584", "issue:45588", "issue:45600", + "issue:45698", "issue:45803", "issue:45987" ], "soft_pairs": [ + "issue:43504|issue:45698", "issue:44734|issue:45406", "issue:44617|issue:45584", "issue:45563|issue:45584", - "issue:43761|issue:44242", "issue:43525|issue:43531", + "issue:43761|issue:44242", "issue:44792|issue:44977", "issue:43931|issue:44060", "issue:44479|issue:44734", "issue:44534|issue:44913", "issue:44190|issue:44964", - "issue:44464|issue:45563", "issue:44792|issue:45600", + "issue:44464|issue:45563", "issue:43827|issue:43994", - "issue:44521|issue:45381", "issue:44855|issue:45588", - "issue:41720|issue:45245", - "issue:45803|issue:45987", + "issue:44521|issue:45381", "issue:35141|issue:42490", + "issue:41720|issue:45245", "issue:43065|issue:44610", + "issue:45803|issue:45987", "issue:44933|issue:45362", "issue:43335|issue:44855", - "issue:43976|issue:44855", "issue:43644|issue:44913", "issue:44077|issue:44079", + "issue:43976|issue:44855", "issue:43550|issue:44190", "issue:43540|issue:44734", - "issue:44737|issue:45803", "issue:45127|issue:45357", - "issue:44112|issue:45600", - "issue:43335|issue:44190" + "issue:44737|issue:45803" ], - "prepared_review_unit_hash": "8fc557f970354796617d3b971403468f552956b844448df6e0b1577a03d2bc17", + "prepared_review_unit_hash": "eebe10e693dc9a620a877971965034125b32849eb8455322da1ec319f8de7c01", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 36105, - "estimated_input_tokens": 9027, - "estimated_eval_tokens": 18310 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36404, + "estimated_input_tokens": 9101, + "estimated_eval_tokens": 18458 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The proposed pairs are mostly unrelated bug reports that only share broad subsystem keywords. I don\u2019t see any soft edge that safely collapses into a single underlying issue. The closest true duplicate theme in the set is the non-persistent-buffer junk regression, with 43644 as the strongest anchor.", - "confidence": 0.91, - "canonical_issue_reason": "issue:43644 is the best canonical anchor: it describes a concrete, repo-wide regression in non-persistent buffer initialization, and it is the earlier, more detailed report compared with the later near-duplicate title in 44534.", + "summary": "This cluster looks heterogeneous: the soft edges are mostly topical lookalikes, not true duplicates. I\u2019d reject every pair and treat issue:44521 as the best single representative only because it has the strongest activity signal, not because it matches the others.", + "confidence": 0.84, + "canonical_issue_reason": "issue:44521 has the strongest discussion/inbound-reference signal in this set, so it is the best representative if a single issue must be chosen, even though the cluster is not a real duplicate group.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43644 is the strongest representative issue because it captures a clear, systemic bug with broad impact and a well-scoped fix target; the rest of the cluster is too heterogeneous to serve as a better canonical.", + "best_issue_reason": "issue:44521 is the most active and well-documented item here, making it the most suitable standalone anchor among otherwise unrelated reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both concern model loading/save behavior, but they describe different failures and code paths." + }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "Different failures: KV-cache continuation indexing in serve vs missing processor state in Gemma4Processor." + "reason": "Both involve transformers serve crashes, but one is KV-cache continuation indexing and the other is a processor attribute error." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "OOM in Sam3Video is unrelated to Whisper empty-transcription failure." + "reason": "Unrelated bugs: video-model OOM versus Whisper empty transcription after token alignment." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "Stale warning for num_return_sequences is unrelated to Whisper\u2019s empty-output regression." + "reason": "Different issues in generation: a stale warning versus a real failure on empty transcription." }, { - "left": "issue:43761", - "right": "issue:44242", + "left": "issue:43525", + "right": "issue:43531", "accept": false, - "reason": "Different models and different bugs: CLIPVision hidden_states regression vs MoE load-balancing loss omission." + "reason": "Different configs and symptoms: missing pad_token_id versus sliding-window handling in Qwen3-MoE." }, { - "left": "issue:43525", - "right": "issue:43531", + "left": "issue:43761", + "right": "issue:44242", "accept": false, - "reason": "Llama4 pad_token_id AttributeError and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "One is a CLIPVisionModel regression; the other is an MoE load-balancing-loss bug." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Janus image-generation test failure is unrelated to Qwen3.5 flash-attention generation behavior." + "reason": "A janus image-generation test failure is not the same bug as Qwen3.5 flash-attention misgeneration." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "Both involve Qwen3 variants, but one is a weight-shape load mismatch and the other is a tied-weights warning bug." + "reason": "Different model-loading failures: weight-shape mismatch versus incorrect tied-weights warning." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Video-input regression across Qwen models is unrelated to serve KV-cache tensor indexing." + "reason": "Both touch generation/runtime behavior, but one is a video-input regression and the other is KV-cache continuation indexing." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk on load is unrelated to GPTNeoX rotary_pct reloading." + "reason": "Non-persistent-buffer initialization and GPTNeoX rotary_pct reload are different root causes." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Local dataset loading failure in a training script is unrelated to Phi-4 multimodal model loading." - }, - { - "left": "issue:44464", - "right": "issue:45563", - "accept": false, - "reason": "Compiled-forward chunked generation inconsistency is unrelated to a stale warning for num_return_sequences." + "reason": "Local dataset loading in a trainer script is unrelated to loading a multimodal model preset." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "Janus test failure and auto_mappings CI breakage are different code paths." + "reason": "A failing janus generation test and stale auto_mappings references are different problems." }, { - "left": "issue:43827", - "right": "issue:43994", + "left": "issue:44464", + "right": "issue:45563", "accept": false, - "reason": "Docs still referencing pipeline() is unrelated to SigLIP2 producing nonsensical outputs." + "reason": "Compiled-forward chunked generation inconsistency is not the same as a stale num_return_sequences warning." }, { - "left": "issue:44521", - "right": "issue:45381", + "left": "issue:43827", + "right": "issue:43994", "accept": false, - "reason": "Both are multimodal, but one is assistant-mask generation and the other is Qwen2.5-VL vision_position_ids." + "reason": "Docs still referencing removed pipeline APIs is unrelated to a broken SigLIP model output." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Python 3.13 indentation/import parsing error is unrelated to flash_attention s_aux=None handling." + "reason": "Python parsing/import failure and flash_attention s_aux handling are separate defects." }, { - "left": "issue:41720", - "right": "issue:45245", + "left": "issue:44521", + "right": "issue:45381", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert is unrelated to category-count overflow." + "reason": "Both involve multimodal chat/video behavior, but the bugs are different: assistant mask generation versus vision_position_ids." }, { - "left": "issue:45803", - "right": "issue:45987", + "left": "issue:35141", + "right": "issue:42490", "accept": false, - "reason": "Bare except swallowing errors in FuyuBatchFeature is unrelated to StaticCache.get_seq_length() returning the wrong type." + "reason": "Token-embedding reinit after resize and a trailing-slash import warning are unrelated." }, { - "left": "issue:35141", - "right": "issue:42490", + "left": "issue:41720", + "right": "issue:45245", "accept": false, - "reason": "Token-embedding resize/post_init reinit bug is unrelated to local-path trailing-slash import warning." + "reason": "Qwen3 auto device-mapping CUDA asserts are unrelated to the category-limit RuntimeError." }, { "left": "issue:43065", "right": "issue:44610", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d is unrelated to OmDet-Turbo processor image-size mismatch." + "reason": "A dummy Conv2d in Sam3PixelDecoder and a processor/model input-size mismatch are different bugs." }, { - "left": "issue:44933", - "right": "issue:45362", + "left": "issue:45803", + "right": "issue:45987", "accept": false, - "reason": "Missing import from image_utils is unrelated to Qwen3.5 chat crashes." + "reason": "Bare except in tensor conversion and StaticCache.get_seq_length() returning the wrong type are unrelated." }, { - "left": "issue:43335", - "right": "issue:44855", + "left": "issue:44933", + "right": "issue:45362", "accept": false, - "reason": "SwitchTransformers sparse-layer construction bug is unrelated to DebertaV2 import parsing on Python 3.13." + "reason": "Missing image_utils import and Qwen3.5 chat crashes are different code paths and symptoms." }, { - "left": "issue:43976", + "left": "issue:43335", "right": "issue:44855", "accept": false, - "reason": "Python-version compatibility issue for Transformers 5.1.0 is unrelated to the DebertaV2 scripting parse error." + "reason": "A sparse-layer config bug and a Python 3.13 import/parsing failure are not the same issue." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "These are different bugs: non-persistent buffer initialization vs config reload losing rotary_pct." + "reason": "The non-persistent-buffer junk regression is distinct from GPTNeoX rotary_pct being lost on reload." }, { "left": "issue:44077", "right": "issue:44079", "accept": false, - "reason": "PatchTSMixer post_init policy and ModelOutput key assignment are unrelated implementation bugs." + "reason": "Optional post_init handling and ModelOutput key assignment are unrelated framework bugs." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "A Python-version compatibility issue is not the same as a DebertaV2Model indentation/parsing regression." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "Bamba torch.compile SDPA failure is unrelated to local dataset loading in a training example." + "reason": "Bamba torch.compile/SDPA failure is unrelated to dataset loading in a training script." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-input ValueError is unrelated to serve KV-cache continuation indexing." - }, - { - "left": "issue:44737", - "right": "issue:45803", - "accept": false, - "reason": "XLNet CPU arange bug is unrelated to a bare except in FuyuBatchFeature." + "reason": "Qwen3OmniMoe video-input ValueError and serve KV-cache indexing crash are different failures." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "LoRA merge collapse with extended vocab is unrelated to wrong visual encoder keys in save_pretrained." - }, - { - "left": "issue:44112", - "right": "issue:45600", - "accept": false, - "reason": "GraniteSpeech CI stale test and auto_mappings reference cleanup are different failures." + "reason": "LoRA merge collapse with extended vocab and incorrect visual-encoder save keys are unrelated." }, { - "left": "issue:43335", - "right": "issue:44190", + "left": "issue:44737", + "right": "issue:45803", "accept": false, - "reason": "SwitchTransformers config bug is unrelated to dataset loading in the training script." + "reason": "XLNet CPU arange device placement and a bare-except swallowing errors are different defects." } ] }, @@ -10088,10 +9973,10 @@ "issue:41628", "issue:41950", "issue:42175", - "issue:42490", "issue:42907", "issue:43065", "issue:43295", + "issue:43335", "issue:43404", "issue:43441", "issue:43479", @@ -10131,245 +10016,252 @@ "issue:46121" ], "soft_pairs": [ + "issue:44112|issue:45600", + "issue:43335|issue:44190", "issue:44462|issue:45907", "issue:46082|issue:46121", "issue:39692|issue:44190", "issue:44938|issue:45542", - "issue:43479|issue:45375", "issue:45276|issue:45584", "issue:43295|issue:45276", + "issue:43479|issue:45375", "issue:36246|issue:41628", "issue:45245|issue:45584", "issue:43650|issue:44220", "issue:44186|issue:44734", - "issue:41950|issue:44016", - "issue:36246|issue:38617", "issue:42175|issue:44279", + "issue:36246|issue:38617", + "issue:41950|issue:44016", "issue:44246|issue:45003", "issue:44683|issue:45446", "issue:43404|issue:45216", - "issue:44016|issue:44220", "issue:38175|issue:43065", + "issue:44016|issue:44220", "issue:43643|issue:44496", - "issue:44734|issue:45632", - "issue:43854|issue:44964", "issue:44964|issue:45600", "issue:44112|issue:44964", + "issue:43854|issue:44964", + "issue:44734|issue:45632", "issue:44561|issue:45632", "issue:43976|issue:44987", "issue:43531|issue:45600", "issue:42907|issue:43441", - "issue:42490|issue:45632", "issue:43295|issue:45335" ], - "prepared_review_unit_hash": "4e79e023105e6bd9466cbd528c214fc02f244305cbe46ef8e4c11ceb59d3513f", + "prepared_review_unit_hash": "a4bd2e9513250d84aaab85b622c21f62e9116bcc531e94792a7bc2c27f755572", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36624, - "estimated_input_tokens": 9156, - "estimated_eval_tokens": 18568 + "soft_pair_count": 31, + "serialized_chars": 36960, + "estimated_input_tokens": 9240, + "estimated_eval_tokens": 18736 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated issues spanning imports, model loading, config validation, serving, and docs/tests. No soft edge looks like the same underlying bug, so all should be rejected. If one issue has to serve as a representative, issue 44964 is the clearest concrete regression report.", - "confidence": 0.92, - "canonical_issue_reason": "No true duplicate hub stands out; issue:44964 is the strongest representative because it reports a concrete, reproducible model-load regression against latest transformers.", + "summary": "The cluster is heterogeneous: it mixes import/API breakages, model-specific regressions, CI/doc issues, and unrelated config/runtime bugs. No soft pair looks like the same concrete fixable bug or change.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 41628 is the clearest broad, user-facing API regression in the set: a top-level import failure that fits the recurring import-breakage theme better than the narrower model/script-specific reports.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44964 is the best single representative: it is specific, user-facing, and clearly describes a reproducible load failure with a named model and version context.", + "best_issue_reason": "Issue 41628 is the best representative because it describes a concrete, high-impact Transformers import regression with a clear repro and wider applicability than the more specialized issues around it.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI-related, but one is a GraniteSpeech test failure and the other is a stale auto_mappings reference; different bugs and different fixes." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "A SwitchTransformers config bug and an image-classification training script dataset-loading problem are unrelated." + }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "Tokenizer file resolution and Hub file listing are different code paths and different bugs." + "reason": "One is AutoTokenizer ignoring tokenizer.json, the other is list_repo_files not forwarding revision; both touch repo/file loading but not the same bug." }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "LlamaConfig head_dim validation and rope-parameter list handling are unrelated config-validation problems." + "reason": "Llama head_dim validation and rope-parameter JSON list handling are distinct config-validation failures." }, { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "Both touch examples/scripts, but one is a SigLIP2 docs example error and the other is local dataset loading in an image-classification script." + "reason": "A SigLIP2 docs example error and a local dataset loading failure in a training script are not the same issue." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 import failure and missing TensorFlow backend with only tensorboard installed are unrelated environment issues." - }, - { - "left": "issue:43479", - "right": "issue:45375", - "accept": false, - "reason": "Both are config-field issues, but they affect different models and different missing/defaulted fields." + "reason": "Python 3.14 load failure and a missing TensorFlow backend when only tensorboard is installed are unrelated environment/backend problems." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Gemma4 embedding resize behavior and Whisper empty transcription after token alignment are unrelated bugs." + "reason": "Gemma4 embedding resize propagation and Whisper empty-transcription generation are different model behaviors." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Processor/tokenizer regression and Gemma4 embedding resize are different model-path problems." + "reason": "Processor/tokenizer regression and Gemma4 token-embedding resize semantics are separate API bugs." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Default multimodal config initialization and missing deepstack_visual_indexes serialization are different config issues." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "Both are import errors, but they concern different symbols and likely different missing exports." + "reason": "Both are import errors, but they are for different symbols/modules and look like separate export problems." }, { "left": "issue:45245", "right": "issue:45584", "accept": false, - "reason": "Category-limit runtime error and Whisper empty-transcription failure are unrelated." + "reason": "Category cardinality limits and Whisper generation on empty output are unrelated." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "A vague data request and an fbank feature extraction bug are not the same underlying issue." + "reason": "A placeholder issue and an fbank feature extraction bug are not related." }, { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "Tokenizer crashes on NER/padding and serving KV-cache indexing are different subsystems and failures." + "reason": "LayoutLMv2 tokenizer crashes and serve/KV-cache tensor indexing failures affect different code paths." }, { - "left": "issue:41950", - "right": "issue:44016", + "left": "issue:42175", + "right": "issue:44279", "accept": false, - "reason": "Video-classification processor lookup and a notebook syntax error are unrelated." + "reason": "Missing TensorFlow in a torch install and a vague dependency issue are not the same concrete bug." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "Different missing imports from different modules; no evidence of a shared root cause." + "reason": "Both are ImportErrors, but they involve different missing symbols in different modules; the fixes would not be merged together." }, { - "left": "issue:42175", - "right": "issue:44279", + "left": "issue:41950", + "right": "issue:44016", "accept": false, - "reason": "General dependency/install complaint is too broad to match the specific TensorFlow-backend packaging bug." + "reason": "A video-classification pipeline bug and a notebook syntax error are unrelated." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import slowness and unsafe sys.modules access may be related only indirectly, but they are not the same concrete bug." + "reason": "Import performance complaints and unsafe sys.modules access may be loosely related, but they are not the same underlying bug." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "A compiled flex_attention failure on new torch versions and a version-check bug for AuxRequest import are distinct issues." + "reason": "Both are flex_attention-related, but one is a compiled-op failure on torch >= 2.9 and the other is a version-check/import gating bug." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Mistral3 lm_head tying and Qwen3.5 save_pretrained checkpoint correctness are different model-specific problems." + "reason": "Weight tying in Mistral3 and incorrect save_pretrained output for Qwen3.5 are different model serialization/training issues." }, { - "left": "issue:44016", - "right": "issue:44220", + "left": "issue:38175", + "right": "issue:43065", "accept": false, - "reason": "Notebook syntax error and feature-extraction failure are unrelated." + "reason": "Zero probabilities in SigLIP2 and a dummy Conv2d in Sam3PixelDecoder are unrelated." }, { - "left": "issue:38175", - "right": "issue:43065", + "left": "issue:44016", + "right": "issue:44220", "accept": false, - "reason": "SigLIP2 zero probabilities and a dummy Conv2d in Sam3PixelDecoder are different model bugs." + "reason": "Notebook syntax and audio feature extraction are different failures." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "Both involve config loading, but one is missing fields with trust_remote_code and the other is unrecognized model/config_type handling." + "reason": "Missing fields from trust_remote_code configs and unrecognized model/type handling are different config-loading problems." }, { - "left": "issue:44734", - "right": "issue:45632", + "left": "issue:44964", + "right": "issue:45600", "accept": false, - "reason": "Serving KV-cache tensor indexing and local trust_remote_code cache-path collision are unrelated." + "reason": "A Phi-4 multimodal load regression and a stale Sam3LiteText mapping reference are separate model/config issues." }, { - "left": "issue:43854", + "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "Both are model-load failures, but for different models and no shared concrete code-path is shown." + "reason": "CI test flakiness in GraniteSpeech and a Phi-4 loading regression are different bugs." }, { - "left": "issue:44964", - "right": "issue:45600", + "left": "issue:43854", + "right": "issue:44964", "accept": false, - "reason": "Phi-4 multimodal loading and broken auto_mappings references to removed Sam3LiteText configs are different breakages." + "reason": "GLM-4.7-Flash unit-test loading failure and Phi-4 multimodal loading failure are model-specific, not the same bug." }, { - "left": "issue:44112", - "right": "issue:44964", + "left": "issue:44734", + "right": "issue:45632", "accept": false, - "reason": "CI test flakiness in GraniteSpeech and Phi-4 multimodal load failure are unrelated." + "reason": "Serve/KV-cache tensor indexing and trust_remote_code cache-path collisions are unrelated." }, { "left": "issue:44561", "right": "issue:45632", "accept": false, - "reason": "trust_remote_code breakage from removing is_torch_fx_available and local cache-path collisions are different bugs." + "reason": "A removed helper breaking trust_remote_code models and a local-model cache collision are different trust_remote_code problems." }, { "left": "issue:43976", "right": "issue:44987", "accept": false, - "reason": "Python-version compatibility and a specific model-loading failure are not the same underlying issue." + "reason": "Python version compatibility and loading a specific model under transformers>=5.1.0 are not the same concrete issue." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and Sam3LiteText mapping references are unrelated." + "reason": "Qwen3-MoE sliding_window behavior and stale auto_mappings references are unrelated." }, { "left": "issue:42907", "right": "issue:43441", "accept": false, - "reason": "Dequantized Ministral/Devstral saving and FlashAttention failure in Ministral-3 are different code paths." - }, - { - "left": "issue:42490", - "right": "issue:45632", - "accept": false, - "reason": "Trailing-slash local-model warning and trust_remote_code cache-path collisions are similar only at a high level; not the same bug." + "reason": "Saving dequantized Ministral/Devstral models and FlashAttention failures are different issues in the same family only by name." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Processor/tokenizer regression and t5gemma embedding resize behavior are unrelated." + "reason": "Processor/tokenizer regression and decoder.embed_tokens resize propagation are distinct API behaviors." } ] }, @@ -10396,10 +10288,12 @@ "issue:38175", "issue:39692", "issue:41950", + "issue:42490", "issue:43295", "issue:43299", "issue:43335", "issue:43381", + "issue:43441", "issue:43479", "issue:43525", "issue:43526", @@ -10407,7 +10301,6 @@ "issue:43650", "issue:43673", "issue:43824", - "issue:43854", "issue:43901", "issue:43976", "issue:44016", @@ -10421,7 +10314,6 @@ "issue:44496", "issue:44521", "issue:44617", - "issue:44683", "issue:44779", "issue:44933", "issue:45020", @@ -10433,27 +10325,29 @@ "issue:45335", "issue:45357", "issue:45405", - "issue:45542", "issue:45584", + "issue:45588", "issue:45593", + "issue:45632", "issue:45759", "issue:45812", "issue:45823", - "issue:46132", "issue:46209" ], "soft_pairs": [ + "issue:42490|issue:45632", "issue:43526|issue:43901", - "issue:44496|issue:45042", "issue:43673|issue:44336", - "issue:44016|issue:44062", + "issue:43441|issue:45588", + "issue:44496|issue:45042", "issue:43335|issue:45593", + "issue:44016|issue:44062", "issue:43824|issue:44933", "issue:44075|issue:44368", "issue:43299|issue:45216", "issue:43381|issue:45593", - "issue:43976|issue:44279", "issue:43901|issue:44496", + "issue:43976|issue:44279", "issue:43650|issue:46209", "issue:44496|issue:45759", "issue:45335|issue:45584", @@ -10463,232 +10357,216 @@ "issue:43299|issue:45357", "issue:44351|issue:45278", "issue:43577|issue:45216", - "issue:41950|issue:45020", "issue:38175|issue:44617", - "issue:41950|issue:45823", + "issue:41950|issue:45020", "issue:39692|issue:43901", "issue:44279|issue:45405", "issue:44779|issue:45812", - "issue:44683|issue:45542", - "issue:43295|issue:44779", - "issue:43976|issue:45405", - "issue:43299|issue:43854", - "issue:43650|issue:46132" + "issue:41950|issue:45823", + "issue:43295|issue:44779" ], - "prepared_review_unit_hash": "b9e92835a2227e37387f4ac88b523cb3013f610ba5b91bb331d9b9897f689fbd", + "prepared_review_unit_hash": "c133a80a7025d31b50e5fbc1e24b25a8cfdcff66b610781a95f1b441f0b9a2a8", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36919, - "estimated_input_tokens": 9230, - "estimated_eval_tokens": 18716 + "soft_pair_count": 29, + "serialized_chars": 36389, + "estimated_input_tokens": 9098, + "estimated_eval_tokens": 18452 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: the soft pairs share superficial themes like regressions, imports, multimodal processors, or config handling, but they do not look like the same underlying bug/change.", - "confidence": 0.79, - "canonical_issue_reason": "If one issue must represent the cluster, issue:45278 is the broadest umbrella because it covers a version-upgrade/import-breakage class that overlaps with several items here. That said, the cluster is too mixed to treat any single issue as a true canonical duplicate target.", + "summary": "This cluster is very heterogeneous: the soft pairs mostly share broad topical language (import errors, multimodal configs, tokenizer regressions, flash-attention, docs) but not the same concrete bug. I would reject all soft edges and treat issue:45278 as only a loose import-regression anchor, not a true duplicate canonical.", + "confidence": 0.9, + "canonical_issue_reason": "No single canonical duplicate set stands out. If forced to anchor the cluster, issue:45278 is the broadest import-regression report, but the rest of the items are too diverse to justify a true canonical duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most generally applicable and user-facing among the set, so it is the best representative issue even though it does not subsume the unrelated tokenizer, multimodal, training, and notebook bugs in this cluster.", + "best_issue_reason": "issue:45278 is the best overall anchor because it is the broadest and most central of the few import-related reports; the other issues are mostly unrelated one-off bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43526", - "right": "issue:43901", + "left": "issue:42490", + "right": "issue:45632", "accept": false, - "reason": "Different surfaces: BeitImageProcessorFast label reduction vs text-classification docs for return_all_scores; not the same bug." + "reason": "Both are path/loading-related, but one is a slash-end warning and the other is a trust_remote_code cache collision; different bugs." }, { - "left": "issue:44496", - "right": "issue:45042", + "left": "issue:43526", + "right": "issue:43901", "accept": false, - "reason": "Model/config loading failure vs PIL image processor dependency regression; unrelated codepaths." + "reason": "Image-label reduction bug vs pipeline docs mismatch; unrelated behavior and code paths." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Generation cache failure during chunked prefill vs ANSI escape handling in loading_report; no shared underlying bug." + "reason": "Generation cache issue in chunked prefill vs ANSI-color output in loading_report; no shared bug." }, { - "left": "issue:44016", - "right": "issue:44062", + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention FlashAttention, but one is a model loading/failure report and the other is an AttributeError on s_aux=None; too different to merge." + }, + { + "left": "issue:44496", + "right": "issue:45042", "accept": false, - "reason": "Notebook syntax error vs tokenizer AddedToken argument duplication; different failure modes entirely." + "reason": "Model config import failure vs PIL image processor requiring torchvision; unrelated." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "Sparse-layer config creation bug vs missing auxiliary losses when denoising is off; not the same change." + "reason": "SwitchTransformers sparse-layer construction bug vs D-FINE auxiliary-loss omission; different models and logic." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error vs tokenizer AddedToken keyword conflict; separate issues." }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Missing Qwen2.5-VL export symbol vs nonexistent image_utils import; different modules and causes." + "reason": "Qwen2.5-VL import failure vs nonexistent image_utils import; both are import problems but not the same module or bug." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "Optimizer hyperparameter handling vs a tie_word_embeddings warning in Qwen3.5 fine-tuning; unrelated." + "reason": "SGD optimizer args ignored vs tie_word_embeddings warning; unrelated." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Qwen3VL model loading regression vs Qwen3.5 save_pretrained serialization regression; different stages and codepaths." + "reason": "Qwen3VL model-loading regression vs Qwen3.5 save_pretrained regression; same family, different code paths and symptoms." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient-checkpointing/eval-mode logic vs D-FINE loss computation; no common underlying defect." + "reason": "Gradient checkpointing in eval mode vs D-FINE auxiliary losses; no overlap beyond being model bugs." }, { - "left": "issue:43976", - "right": "issue:44279", + "left": "issue:43901", + "right": "issue:44496", "accept": false, - "reason": "Python-version compatibility regression vs vague dependency issue; too broad and not the same bug." + "reason": "Pipeline docs note vs unrecognized model/config loading error; unrelated." }, { - "left": "issue:43901", - "right": "issue:44496", + "left": "issue:43976", + "right": "issue:44279", "accept": false, - "reason": "Outdated docs around return_all_scores vs an actual model-loading error; different artifacts and fixes." + "reason": "Python version compatibility problem vs generic dependency issue; not the same bug." }, { "left": "issue:43650", "right": "issue:46209", "accept": false, - "reason": "Placeholder issue text vs FP8 inference on B200 hardware; clearly unrelated." + "reason": "Placeholder/data request vs FP8 inference failure; unrelated." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "Unrecognized multimodal model loading vs AutoModelForCausalLM not unwrapping text_config; both config-related but distinct codepaths." + "reason": "Unrecognized model config vs AutoModelForCausalLM not unwrapping text_config; distinct multimodal-loading problems." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Token-embedding resize bug vs Whisper empty-transcription generation failure after token alignment; unrelated." + "reason": "Token embedding resize bug vs Whisper empty-transcription generation failure; different subsystems." }, { "left": "issue:43479", "right": "issue:43525", "accept": false, - "reason": "Default sub-config initialization in Phi4MultimodalConfig vs missing pad_token_id attribute in Llama4Config; different config bugs." + "reason": "Phi4Multimodal default-config initialization vs Llama4 missing pad_token_id; both config issues but not the same defect." }, { "left": "issue:44242", "right": "issue:45200", "accept": false, - "reason": "Router load-balancing loss omission vs Gemma4 mm_token_type_ids defaulting; separate multimodal/training behaviors." + "reason": "MoE load-balancing loss gating bug vs Gemma4 mm_token_type_ids defaulting; different training behaviors." }, { "left": "issue:44521", "right": "issue:45325", "accept": false, - "reason": "assistant mask generation in chat templating vs rope-index temporal scaling; both multimodal but different mechanisms." + "reason": "All-zero assistant mask bug vs rope-index temporal scaling bug; both multimodal but not the same underlying change." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "Qwen3VL loading issue vs Qwen3.5 save_pretrained visual-key regression; different model families and serialization paths." + "reason": "Qwen3.5/Qwen3VL load regression vs save_pretrained visual-key regression; related family, different failures." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "Specific HybridCache import failure vs broad import errors after upgrade; related theme but not the same underlying bug." + "reason": "HybridCache import error may be one symptom of broader import regressions, but the titles are too broad/specific to confirm the same underlying bug." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "dtype propagation on Blip2 load vs Qwen3.5 checkpoint-saving regression; distinct model-specific issues." - }, - { - "left": "issue:41950", - "right": "issue:45020", - "accept": false, - "reason": "Video-classification pipeline choosing image processors vs remote_code model breakage; not the same codepath." + "reason": "BLIP2 dtype propagation bug vs Qwen3.5 checkpoint save regression; unrelated." }, { "left": "issue:38175", "right": "issue:44617", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 vs CUDA OOM in Sam3Video; different symptom and root cause." + "reason": "SigLIP2 zero-probabilities issue vs Sam3Video OOM; completely different problems." }, { "left": "issue:41950", - "right": "issue:45823", + "right": "issue:45020", "accept": false, - "reason": "Pipeline processor lookup bug vs Gemma4 device placement mismatch; unrelated bugs." + "reason": "Video-classification pipeline processor lookup bug vs remote_code breakage in recent versions; not the same issue." }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "SigLIP2 docs example with model/processor mismatch and quantization failure vs a different docs mismatch about return_all_scores." + "reason": "SigLIP2 docs example errors vs text-classification docs mismatch; both docs-related but different examples and failures." }, { "left": "issue:44279", "right": "issue:45405", "accept": false, - "reason": "Generic dependency issue vs PEFT minimum-version bump to an unreleased release; not the same defect." + "reason": "Generic dependency issue vs MIN_PEFT_VERSION bump/release problem; unrelated." }, { "left": "issue:44779", "right": "issue:45812", "accept": false, - "reason": "Tokenizer regression for Deepseek models vs Granite tokenizer regression; same broad area, but different models and likely different fixes." + "reason": "Deepseek tokenizer regression vs Granite tokenizer regression; both tokenizer bugs, but different model families and likely different fixes." }, { - "left": "issue:44683", - "right": "issue:45542", + "left": "issue:41950", + "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on torch>=2.9 vs optional backend detection for tensorboard/TensorFlow; unrelated." + "reason": "Video-classification processor lookup bug vs Gemma4 PLE device mismatch; unrelated." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Processor/tokenizer API regression in custom model code vs Deepseek tokenization correctness regression; different failure modes." - }, - { - "left": "issue:43976", - "right": "issue:45405", - "accept": false, - "reason": "Python compatibility problem vs unreleased PEFT version bump; unrelated dependency concerns." - }, - { - "left": "issue:43299", - "right": "issue:43854", - "accept": false, - "reason": "Qwen3VL MoE loading regression vs GLM-4.7-Flash loading in tests; different model families and loader paths." - }, - { - "left": "issue:43650", - "right": "issue:46132", - "accept": false, - "reason": "Empty/placeholder issue vs AttentionInterface.register behavior change; no duplicate relationship." + "reason": "Processor/tokenizer regression with multimodal inputs vs Deepseek tokenizer wrong results; different code paths and model classes." } ] }, @@ -10717,17 +10595,16 @@ "issue:42994", "issue:43065", "issue:43262", + "issue:43299", "issue:43317", "issue:43441", "issue:43450", - "issue:43525", + "issue:43650", "issue:43653", "issue:43673", "issue:43723", "issue:43761", - "issue:43784", - "issue:43824", - "issue:43881", + "issue:43854", "issue:43901", "issue:43976", "issue:43994", @@ -10742,7 +10619,6 @@ "issue:44462", "issue:44466", "issue:44521", - "issue:44617", "issue:44623", "issue:44671", "issue:44683", @@ -10752,6 +10628,7 @@ "issue:45081", "issue:45276", "issue:45397", + "issue:45405", "issue:45542", "issue:45593", "issue:45600", @@ -10759,241 +10636,235 @@ "issue:45712", "issue:45823", "issue:45834", - "issue:45907" + "issue:45907", + "issue:46132" ], "soft_pairs": [ + "issue:44683|issue:45542", + "issue:43299|issue:43854", + "issue:43976|issue:45405", + "issue:43650|issue:46132", "issue:44462|issue:45081", "issue:43450|issue:44279", - "issue:44016|issue:44291", "issue:44297|issue:44462", - "issue:43441|issue:45834", + "issue:44016|issue:44291", "issue:45081|issue:45907", "issue:44466|issue:45276", + "issue:43441|issue:45834", "issue:45600|issue:45712", "issue:42994|issue:43317", - "issue:33290|issue:45656", "issue:43653|issue:45593", - "issue:44623|issue:44749", + "issue:33290|issue:45656", "issue:44623|issue:45397", + "issue:44623|issue:44749", "issue:43065|issue:44898", "issue:44261|issue:44263", - "issue:43441|issue:44683", "issue:44361|issue:45593", + "issue:43441|issue:44683", "issue:44016|issue:45823", + "issue:44671|issue:44779", "issue:42175|issue:43976", "issue:43723|issue:44162", - "issue:44671|issue:44779", - "issue:43262|issue:44521", "issue:43901|issue:43994", - "issue:43673|issue:43761", - "issue:43784|issue:45542", - "issue:43824|issue:43976", - "issue:43450|issue:44617", - "issue:44297|issue:45397", - "issue:43525|issue:43881" + "issue:43262|issue:44521", + "issue:43673|issue:43761" ], - "prepared_review_unit_hash": "c4885effb66f2cf40f2432dd719626865cad5df061840c83a793500b67ba4700", + "prepared_review_unit_hash": "2d226b252c2693489c5d52387b60d993a4440c3bae14104c035adb38cd540d75", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36319, - "estimated_input_tokens": 9080, - "estimated_eval_tokens": 18416 + "soft_pair_count": 28, + "serialized_chars": 36010, + "estimated_input_tokens": 9003, + "estimated_eval_tokens": 18262 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very heterogeneous: the soft-similarity pairs are mostly topical overlaps (tokenizers, attention, deepspeed, processors, CI) rather than true duplicates. I would not merge any of the suggested pairs.", - "confidence": 0.88, - "canonical_issue_reason": "Issue #44521 is the strongest anchor candidate: it is active, user-facing, has multiple inbound references, and describes a clear multimodal processor regression with broad impact.", + "summary": "The items do not form a true duplicate cluster; they span many unrelated bugs across tokenizers, model loading, processors, docs, and packaging. I would treat issue 44521 as the best representative issue, but none of the soft-edge pairs look like the same underlying bug/change.", + "confidence": 0.92, + "canonical_issue_reason": "Best representative issue here: it is a concrete, reproducible API regression in `apply_chat_template`, has multiple inbound references, and is broader/more actionable than the many model-specific or unrelated reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #44521 is the best duplicate-target candidate among the issues because it is well-scoped, current, and likely to attract related reports if any exist.", + "best_issue_reason": "Issue 44521 is the strongest standalone issue: clear symptom, good downstream impact, active discussion, and a specific code-path to fix.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Different subsystems and failure modes: compiled flex attention on new torch vs missing TensorFlow backend installation." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Both are model-loading problems, but they concern different models and likely different code paths." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Unrelated packaging/version regressions: Python support vs an over-bumped PEFT minimum version." + }, + { + "left": "issue:43650", + "right": "issue:46132", + "accept": false, + "reason": "No shared bug: one is a placeholder/data issue, the other is a behavioral change in attention registration." + }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer-loading bugs, but one is AutoTokenizer ignoring tokenizer.json and the other is a Mistral regex patch crash; different failure paths." + "reason": "Both involve tokenizers, but one is AutoTokenizer precedence for tokenizer.json and the other is a Mistral regex-patch crash." }, { "left": "issue:43450", "right": "issue:44279", "accept": false, - "reason": "Unrelated issues: batched video processor shape bug vs a generic dependency problem." - }, - { - "left": "issue:44016", - "right": "issue:44291", - "accept": false, - "reason": "Notebook syntax error and init_empty_weights TypeError are different problems with no shared code path." + "reason": "Completely unrelated: video processor batching shape bug vs a general dependency issue." }, { "left": "issue:44297", "right": "issue:44462", "accept": false, - "reason": "Both touch tokenizer save/load behavior, but one is tokenizer_class serialization mismatch and the other is repository tokenizer file selection." + "reason": "Different bugs in tokenizer serialization vs tokenizer loading precedence; related area, not the same underlying defect." }, { - "left": "issue:43441", - "right": "issue:45834", + "left": "issue:44016", + "right": "issue:44291", "accept": false, - "reason": "Both involve model-specific failures, but Ministral FlashAttention and Kosmos2.5 OCR index errors are unrelated bugs." + "reason": "Notebook syntax error and init_empty_weights TypeError are unrelated." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Mistral regex crash and list_repo_files revision forwarding are unrelated repository/tokenizer issues." + "reason": "Different issues: Mistral tokenizer regex patch crash vs list_repo_files revision forwarding in CI." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Different regressions: tied-weight serialization vs Gemma4 resize_token_embeddings propagation." + "reason": "Both touch Gemma4, but one is tied-weight serialization and the other is resize_token_embeddings behavior." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Different models and symptoms: FlashAttention failure vs OCR index error." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Both are cleanup/CI-related, but they concern different files and different dummy/reference issues." + "reason": "Both affect repository maintenance/CI, but they are separate file/reference problems with no shared code path." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Quantized model saving failure and dequantized model loading with offload are separate lifecycle bugs." + "reason": "Both touch quantization/loading, but one is saving a quantized model and the other is dequantized offload loading." }, { - "left": "issue:33290", - "right": "issue:45656", + "left": "issue:43653", + "right": "issue:45593", "accept": false, - "reason": "DeepSpeed-related, but adafactor OOM and optimizer.step being called twice are distinct symptoms and likely different code paths." + "reason": "Tokenizer special-token decoding bug vs D-FINE auxiliary-loss behavior; unrelated." }, { - "left": "issue:43653", - "right": "issue:45593", + "left": "issue:33290", + "right": "issue:45656", "accept": false, - "reason": "BigBirdTokenizer special-token registration and D-FINE auxiliary-loss handling are unrelated." + "reason": "Both mention DeepSpeed/optimizer, but the concrete bugs differ: Adafactor OOM vs optimizer.step being called twice." }, { "left": "issue:44623", - "right": "issue:44749", + "right": "issue:45397", "accept": false, - "reason": "Processor save_pretrained file omission and a performance regression after upgrade are not the same bug." + "reason": "Processor save-pretrained file omission and Gemma-4 zero3 from_pretrained are different regressions." }, { "left": "issue:44623", - "right": "issue:45397", + "right": "issue:44749", "accept": false, - "reason": "Processor save_pretrained missing files and Gemma-4 zero3 from_pretrained are unrelated." + "reason": "No match: missing processor save files vs performance regression after upgrade." }, { "left": "issue:43065", "right": "issue:44898", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to Perceiver non-default-resolution failures." + "reason": "Different vision bugs in different models: dummy Conv2d placeholder vs Perceiver interpolation failure." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "These are different implementation bugs in different models: missing rms_norm_eps vs torch.split return handling." + "reason": "Different model components and symptoms: missing rms_norm_eps precision drift vs torch.split return-value handling." }, { - "left": "issue:43441", - "right": "issue:44683", + "left": "issue:44361", + "right": "issue:45593", "accept": false, - "reason": "Both mention attention, but one is FlashAttention on a specific model and the other is compiled flex_attention on torch >= 2.9." + "reason": "Tokenizer attribute error vs D-FINE loss behavior; unrelated." }, { - "left": "issue:44361", - "right": "issue:45593", + "left": "issue:43441", + "right": "issue:44683", "accept": false, - "reason": "MLukeTokenizer task attribute error and D-FINE auxiliary-loss behavior are unrelated." + "reason": "Different feature failures: model-specific FlashAttention regression vs compiled flex_attention failing on torch 2.9+." }, { "left": "issue:44016", "right": "issue:45823", "accept": false, - "reason": "Notebook syntax error and Gemma4 PLE device mismatch are different issues." + "reason": "Notebook syntax error and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Both are model-specific wrong-results reports, but they concern different tokenizers/models and distinct regressions." }, { "left": "issue:42175", "right": "issue:43976", "accept": false, - "reason": "Packaging/backends issue and Python version compatibility issue are not duplicates." + "reason": "Packaging/backend dependency mismatch vs Python version support breakage; not the same bug." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Tokenizer loading in v5 and ESM2 breakage are separate model/framework failures." + "reason": "AutoTokenizer loading regression and ESM2 breakage are different issues." }, { - "left": "issue:44671", - "right": "issue:44779", + "left": "issue:43901", + "right": "issue:43994", "accept": false, - "reason": "CamemBERT masked-LM predictions and Deepseek tokenizer output are different regressions." + "reason": "Docs mismatch for a pipeline option vs incorrect model/pipeline outputs; not the same underlying change." }, { "left": "issue:43262", "right": "issue:44521", "accept": false, - "reason": "Both are processor/chat-template related, but one is default sampling rate and the other is assistant mask generation for multimodal inputs." - }, - { - "left": "issue:43901", - "right": "issue:43994", - "accept": false, - "reason": "Pipeline docs mismatch and SigLIP2 nonsensical inference are unrelated." + "reason": "Same API name, but different modalities and symptoms: default 16kHz audio processing vs multimodal assistant mask generation." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache/chunked prefill and CLIPVisionModel hidden_states are different regression classes." - }, - { - "left": "issue:43784", - "right": "issue:45542", - "accept": false, - "reason": "Importing sentence-transformers with missing nn and tensorboard/TensorFlow backend availability 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 issues." - }, - { - "left": "issue:43450", - "right": "issue:44617", - "accept": false, - "reason": "Video processor batching shape bug and Sam3Video CUDA OOM are not the same underlying defect." - }, - { - "left": "issue:44297", - "right": "issue:45397", - "accept": false, - "reason": "Tokenizer save metadata mismatch and Gemma-4 zero3 loading are unrelated." - }, - { - "left": "issue:43525", - "right": "issue:43881", - "accept": false, - "reason": "Llama4Config pad_token_id AttributeError and glm-4v-9b loading failure are different model-specific bugs." + "reason": "Generation cache regression and CLIPVision hidden_states bug are unrelated." } ] }, @@ -11023,23 +10894,26 @@ "issue:42491", "issue:42757", "issue:42886", - "issue:43262", "issue:43299", "issue:43352", + "issue:43450", "issue:43525", "issue:43526", "issue:43531", "issue:43596", "issue:43643", "issue:43673", + "issue:43784", + "issue:43824", + "issue:43881", "issue:43901", "issue:43976", - "issue:44162", "issue:44188", "issue:44220", "issue:44242", "issue:44246", "issue:44295", + "issue:44297", "issue:44336", "issue:44351", "issue:44373", @@ -11047,34 +10921,36 @@ "issue:44496", "issue:44561", "issue:44589", + "issue:44617", "issue:44749", "issue:44779", "issue:44857", "issue:44871", - "issue:44877", "issue:45003", - "issue:45200", "issue:45250", "issue:45292", "issue:45362", - "issue:45375", "issue:45397", - "issue:45412", + "issue:45542", "issue:45561", "issue:45563", "issue:45593", - "issue:46123", "issue:46132" ], "soft_pairs": [ + "issue:43784|issue:45542", + "issue:43824|issue:43976", + "issue:43450|issue:44617", + "issue:44297|issue:45397", + "issue:43525|issue:43881", "issue:43901|issue:44246", "issue:43525|issue:43643", "issue:44188|issue:44336", "issue:41628|issue:43976", "issue:44589|issue:45362", + "issue:44336|issue:44373", "issue:42491|issue:43299", "issue:45561|issue:45563", - "issue:44336|issue:44373", "issue:44295|issue:45397", "issue:44246|issue:44496", "issue:44242|issue:45593", @@ -11088,248 +10964,233 @@ "issue:44246|issue:44871", "issue:45003|issue:45561", "issue:43673|issue:44561", - "issue:44242|issue:44857", "issue:44779|issue:45292", "issue:36246|issue:43596", - "issue:43352|issue:45250", "issue:44295|issue:44749", - "issue:44162|issue:45200", - "issue:43901|issue:45003", - "issue:45412|issue:46123", - "issue:44496|issue:45003", - "issue:44877|issue:45375", - "issue:43262|issue:44246" + "issue:44242|issue:44857", + "issue:43352|issue:45250" ], - "prepared_review_unit_hash": "873e4f51a09dcb2a564583796115c1877a2309c823ecd52bb1f6a0dfd60925ed", + "prepared_review_unit_hash": "ea79cb538e4fdc291f985295c7e0e9840a170a2efb34e5f2bafba8e320daa6e8", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 37081, - "estimated_input_tokens": 9271, - "estimated_eval_tokens": 18798 + "soft_pair_count": 31, + "serialized_chars": 36692, + "estimated_input_tokens": 9173, + "estimated_eval_tokens": 18602 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are distinct bug reports across different model families and subsystems. I found one plausible duplicate pair around strict config handling; the rest are too semantically different to merge as duplicates.", - "confidence": 0.79, - "canonical_issue_reason": "issue:44877 is the broadest, most root-cause-oriented report in the only near-duplicate pair: it describes strict config loading failing on model configs, which also covers the more specific field-dropping case in issue:45375.", + "summary": "This cluster is mostly loose semantic similarity rather than true duplicates. I rejected all soft edges, and there are no PRs in the set. If forced to pick a representative, issue:44246 is the least-bad broad node, but it is only a weak canonical.", + "confidence": 0.66, + "canonical_issue_reason": "issue:44246 is the broadest, most generic report in the cluster and sits on multiple soft-similarity edges, so it is the best available representative despite the cluster being heterogeneous.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44877 best represents the shared strict-config regression because it is model-agnostic, clearly describes the loading failure, and can subsume the narrower config-field complaint.", + "best_issue_reason": "Same as canonical: it is the most general issue and the closest thing to a central triage anchor here, though not a true duplicate hub.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Both are import/backend failures, but the symbols and failure modes differ (`nn` import vs TensorBoard/TensorFlow backend)." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Missing Qwen2.5-VL export vs Python version compatibility; same package, different bugs." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Video processor shape bug vs SAM video OOM; different symptoms and code paths." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save config mismatch is unrelated to Gemma-4 ZeRO3 loading." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Missing `pad_token_id` on Llama4Config is unrelated to glm-4v-9b loading failure." + }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Docs wording about return_all_scores and an import-slowdown report are different problems in different code paths." + "reason": "Docs mismatch for `return_all_scores` is not the same as intermittent import slowness." }, { "left": "issue:43525", "right": "issue:43643", "accept": false, - "reason": "Llama4 pad_token_id missing and trust_remote_code dropping fields are both config-related, but they are not the same bug." + "reason": "Llama4 config attribute error and `trust_remote_code` missing fields are different failures." }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "Attention-kernel divergence under torch.compile is unrelated to ANSI codes emitted by loading_report." + "reason": "Attention-kernel divergence and ANSI output in loading_report are unrelated." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Missing AutoImageProcessor import and Python 3.9+ incompatibility are separate failures with no shared concrete code path." + "reason": "AutoImageProcessor import error is not the same as the Python 3.9/3.10 support issue." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Different runtime errors in different model/chat paths; no evidence of the same underlying defect." + "reason": "Float8 storage lookup failure and Qwen3.5 chat crash are different model/runtime problems." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI code leakage and a wrong docstring do not share the same underlying bug." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both involve Qwen/MoE models, but one is LoRA compatibility and the other is a Qwen3VL loading regression." + "reason": "Both mention Qwen3-MoE, but one is hf4.x/hf5.x LoRA compatibility and the other is a v5 loading regression." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "A pytest-xdist race in testing utils is unrelated to a stale generation warning." - }, - { - "left": "issue:44336", - "right": "issue:44373", - "accept": false, - "reason": "ANSI output leakage and a wrong docstring for position_ids are unrelated." + "reason": "A pytest-xdist race on `captured_info.txt` is unrelated to a stale `num_return_sequences` warning." }, { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Reading position_ids after buffer registration is not the same as a gemma-4 zero3 from_pretrained failure." + "reason": "Buffered `position_ids` access and Gemma-4 ZeRO3 loading are different code paths." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "Import slowness and an unrecognized model/config loading error are different issues." + "reason": "Import performance issue is unrelated to an unrecognized model/config error." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Different auxiliary-loss bugs in different model families; they are not the same concrete change." + "reason": "These are both auxiliary-loss omissions, but for different models and different gating flags; not one concrete bug." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Both mention Qwen models, but sliding_window handling and a transformers chat crash are distinct failures." + "reason": "Qwen3-MoE `sliding_window` behavior and Qwen3.5 chat crashes are not the same issue." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Video-classification pipeline image-processor lookup and audio fbank extraction are unrelated." + "reason": "Video-classification image-processor lookup and `_torch_extract_fbank_features()` are unrelated." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast label reduction and unsafe sys.modules access are unrelated bugs." + "reason": "`reduce_labels` returning one label and unsafe `sys.modules` access are different bugs." }, { "left": "issue:42886", "right": "issue:46132", "accept": false, - "reason": "Offline cache loading and AttentionInterface.register behavior changes are different code paths." + "reason": "Offline cache loading and `AttentionInterface.register` behavior change are unrelated." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols from different packages and with different causes." + "reason": "Both are import errors, but they involve different missing symbols and different libraries." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Missing Qwen2_5_VLImageProcessor and missing HybridCache exports are separate symbol-export issues." + "reason": "Different missing `transformers` exports; same package, not the same bug." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "Terminal ANSI generation and GLM-5 RoPE implementation are unrelated." + "reason": "ANSI formatting leakage and GLM-5 RoPE discussion are unrelated." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import performance regression and Gemma eos_token_id mismatch do not share the same bug." + "reason": "Import slowness and Gemma eos_token_id inconsistency are different problems." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils is unrelated to the xdist captured_info.txt race." + "reason": "Unsafe `sys.modules` access is unrelated to the xdist file race." }, { "left": "issue:43673", "right": "issue:44561", "accept": false, - "reason": "Both are v5 regressions, but one is GenerationMixin cache/chunked_prefill and the other is removal of is_torch_fx_available for trust_remote_code." - }, - { - "left": "issue:44242", - "right": "issue:44857", - "accept": false, - "reason": "Missing load-balancing loss and a float16 AMP crash in LwDetrImageLoss are different defects." + "reason": "Both touch v5 remote-code breakage, but the concrete defects differ (`GenerationMixin` cache vs removed helper)." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Tokenizer correctness in Deepseek and resize_token_embeddings propagation are unrelated." + "reason": "Deepseek tokenizer correctness regression and `resize_token_embeddings` not updating output embeddings are different bugs." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "Missing image-processor import and a DeepSpeed zero3 BertModel IndexError are unrelated." - }, - { - "left": "issue:43352", - "right": "issue:45250", - "accept": false, - "reason": "A model-specific Flash Attention 2 unsupported error is not the same as the generic feature/topic issue." + "reason": "Qwen2.5-VL import error and BertModel/deepspeed IndexError are unrelated." }, { "left": "issue:44295", "right": "issue:44749", "accept": false, - "reason": "A position_ids buffer error and a Chinese report about slower filtering after upgrade are unrelated." - }, - { - "left": "issue:44162", - "right": "issue:45200", - "accept": false, - "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaulting are different model-specific issues." - }, - { - "left": "issue:43901", - "right": "issue:45003", - "accept": false, - "reason": "A pipeline doc mismatch is not the same as unsafe sys.modules access." - }, - { - "left": "issue:45412", - "right": "issue:46123", - "accept": false, - "reason": "RT-DETR memory lifecycle and MaskGenerationPipeline partial-batch dropping are different bugs." + "reason": "Buffered `position_ids` handling and slower filtering after upgrade are not the same issue." }, { - "left": "issue:44496", - "right": "issue:45003", + "left": "issue:44242", + "right": "issue:44857", "accept": false, - "reason": "Model config recognition failure and sys.modules access misuse are distinct root causes." - }, - { - "left": "issue:44877", - "right": "issue:45375", - "accept": true, - "reason": "Both reports point to the same strict-config regression: model config fields are being rejected or dropped during loading, with 45375 being a specific instance of the broader 44877 failure mode." + "reason": "MoE loss gating and an AMP crash in LwDetrImageLoss are unrelated." }, { - "left": "issue:43262", - "right": "issue:44246", + "left": "issue:43352", + "right": "issue:45250", "accept": false, - "reason": "Audio sampling-rate defaults and import-time slowness are unrelated." + "reason": "A model-specific Flash Attention 2 support gap is not the same as the generic Flash Attention 2 issue." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The single accepted soft-edge pair, issue:44877 and issue:45375, is plausibly the same strict-config loading regression based on the titles alone. The remaining soft-edge rejections also read as cautious and evidence-based, without overclaiming duplicates." - }, - "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 }, @@ -11350,30 +11211,29 @@ "issue:43065", "issue:43232", "issue:43257", + "issue:43262", "issue:43408", "issue:43475", "issue:43540", "issue:43653", - "issue:43673", + "issue:43901", "issue:43931", "issue:44060", "issue:44077", "issue:44162", "issue:44188", - "issue:44297", + "issue:44246", "issue:44336", "issue:44368", "issue:44466", "issue:44479", "issue:44483", - "issue:44530", - "issue:44623", - "issue:44734", + "issue:44496", "issue:44877", - "issue:44933", "issue:44977", "issue:44987", "issue:44995", + "issue:45003", "issue:45137", "issue:45200", "issue:45229", @@ -11381,6 +11241,7 @@ "issue:45335", "issue:45375", "issue:45397", + "issue:45412", "issue:45459", "issue:45563", "issue:45584", @@ -11396,238 +11257,245 @@ "issue:46202" ], "soft_pairs": [ + "issue:44162|issue:45200", + "issue:43901|issue:45003", + "issue:44496|issue:45003", + "issue:45412|issue:46123", + "issue:44877|issue:45375", + "issue:43262|issue:44246", "issue:45292|issue:45584", "issue:43257|issue:45137", - "issue:45987|issue:46082", - "issue:45375|issue:45759", "issue:45292|issue:45701", + "issue:45375|issue:45759", + "issue:45987|issue:46082", "issue:43232|issue:44336", "issue:44987|issue:45459", "issue:44479|issue:46170", "issue:43065|issue:43408", - "issue:43475|issue:43540", "issue:44995|issue:45200", + "issue:43475|issue:43540", "issue:44466|issue:45335", - "issue:42886|issue:45229", "issue:45397|issue:45656", + "issue:42886|issue:45229", "issue:44483|issue:46202", - "issue:44188|issue:46202", "issue:44060|issue:44077", + "issue:44188|issue:46202", "issue:44995|issue:45588", + "issue:45137|issue:45600", "issue:43653|issue:44162", "issue:44368|issue:44977", - "issue:45137|issue:45600", "issue:44877|issue:46170", "issue:44368|issue:45563", - "issue:43673|issue:44734", - "issue:43931|issue:44368", - "issue:44530|issue:46170", - "issue:43065|issue:46123", - "issue:44933|issue:45459", - "issue:44297|issue:44623" + "issue:43931|issue:44368" ], - "prepared_review_unit_hash": "5f779d072cbd0f23229366b9a90a089a57a47b34209a3e944d8672a578ec7227", + "prepared_review_unit_hash": "415d1b11b004e591ae623baad14771715efc3a5d173196de66df367e745cf206", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 36398, - "estimated_input_tokens": 9100, - "estimated_eval_tokens": 18456 + "soft_pair_count": 30, + "serialized_chars": 36748, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very heterogeneous: most items are unrelated model-, tokenizer-, config-, or serving-specific bugs that only share broad Transformers/deepspeed vocabulary. None of the proposed soft edges look like true duplicates.", - "confidence": 0.92, - "canonical_issue_reason": "Issue 45292 is the best anchor only by being the most general, self-contained API regression in the set; it is concrete, reproducible, and less model-specific than most others.", + "summary": "The cluster is highly heterogeneous: the soft pairs mostly share only broad subsystem vocabulary (embeddings, config loading, deepspeed, video, tokenizers) rather than the same concrete bug. I reject all soft edges as non-duplicates.", + "confidence": 0.94, + "canonical_issue_reason": "If forced to choose one representative, issue 44466 is the closest hub because several nearby items revolve around tied-weight / embedding behavior. That said, the cluster does not form a single coherent duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45292 is the strongest standalone issue to keep because it describes a clear core-library behavior gap rather than a niche model-family or environment-specific failure.", + "best_issue_reason": "Issue 44466 is the most concrete and broadly scoped among the recurring embedding/tied-weight reports, so it is the best single issue to anchor from this mixed set.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "Different models and failures: ESM2 is broken vs Gemma4 mm_token_type_ids defaulting." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch in a pipeline vs unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model/config loading vs sys.modules access bug; unrelated code paths." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "Memory leak on RT-DETR deletion vs final-batch pipeline flag bug." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Both involve config strictness, but different models and different missing fields." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate default vs slow import; no shared bug." + }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Resize-token-embedding behavior is unrelated to Whisper empty-transcription generation after align_special_tokens." + "reason": "Resize-token-embeddings/output head sync vs Whisper empty transcription after align_special_tokens." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Both mention DeepSpeed, but one is weight conversion during load and the other is a ZeRO3 deque IndexError; different failures." + "reason": "DeepSpeed-related, but one is weight conversion and the other is a zero3 deque crash." }, { - "left": "issue:45987", - "right": "issue:46082", + "left": "issue:45292", + "right": "issue:45701", "accept": false, - "reason": "Cache sequence-length return type and LlamaConfig head_dim validation are unrelated bugs." + "reason": "Embedding resize bug vs tokenization changing across versions; too broad to merge." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Missing config field in Qwen3.5 MoE vision config is not the same as AutoModelForCausalLM not unwrapping text_config." + "reason": "Missing config field under @strict vs from_config not unwrapping text_config." }, { - "left": "issue:45292", - "right": "issue:45701", + "left": "issue:45987", + "right": "issue:46082", "accept": false, - "reason": "Embedding resize/output sync bug is unrelated to a version-dependent tokenization change." + "reason": "Cache-length return type bug vs LlamaConfig head_dim validation; unrelated." }, { "left": "issue:43232", "right": "issue:44336", "accept": false, - "reason": "Generation kwargs handling after sync_gpus is unrelated to ANSI formatting in loading_report." + "reason": "Generation kwargs/sync_gpus bug vs ANSI escape output in loading_report." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Both involve loading errors, but one is a model/package load failure and the other is tokenizer error masking when protobuf is absent." + "reason": "Model load failure in v5.1+ vs tokenizer error masking when protobuf is absent." }, { "left": "issue:44479", "right": "issue:46170", "accept": false, - "reason": "Video-input regression in Qwen models is unrelated to Windows ROCm import/type errors from torch.distributed." + "reason": "Video-input regression vs Windows/ROCm distributed import failure; different stacks." }, { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "A dummy Conv2d in Sam3PixelDecoder is not the same as a model-type warning for sam3_video vs sam3_tracker." - }, - { - "left": "issue:43475", - "right": "issue:43540", - "accept": false, - "reason": "Missing fpn_position_embeddings in SAM3 vision output is unrelated to Qwen3OmniMoe video input validation." + "reason": "SAM3 dummy Conv2d implementation vs model-type mismatch warning; not the same defect." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "Stale indexer cache crash and required mm_token_type_ids for Gemma 4 fine-tuning are different code paths." + "reason": "Stale cache crash in GlmMoeDsa vs Gemma4 token-type defaults." }, { - "left": "issue:44466", - "right": "issue:45335", + "left": "issue:43475", + "right": "issue:43540", "accept": false, - "reason": "Tied-weight serialization inconsistency by device is not the same as resize_token_embeddings failing to update decoder.embed_tokens." + "reason": "SAM3 vision encoder attribute error vs Qwen3OmniMoe video processing ValueError." }, { - "left": "issue:42886", - "right": "issue:45229", + "left": "issue:44466", + "right": "issue:45335", "accept": false, - "reason": "Offline tokenizer cache loading and Gemma4 multi-GPU CUDA OOM are unrelated." + "reason": "Both touch embeddings, but one is serialization inconsistency and the other is resize_token_embeddings propagation." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma-4 zero3 from_pretrained and optimizer stepping twice under deepspeed are different bugs." + "reason": "Gemma-4 zero3 loading vs duplicate optimizer.step under deepspeed; distinct bugs." }, { - "left": "issue:44483", - "right": "issue:46202", + "left": "issue:42886", + "right": "issue:45229", "accept": false, - "reason": "Chat/completions request rejection is unrelated to Gemma pipeline roundtrip failure on \u2581." + "reason": "Offline tokenizer cache load failure vs Gemma4 multi-GPU CUDA OOM." }, { - "left": "issue:44188", + "left": "issue:44483", "right": "issue:46202", "accept": false, - "reason": "torch.compile attention-kernel divergence is unrelated to Gemma token roundtrip issues." + "reason": "Chat/completions API rejection vs Gemma pipeline roundtrip on \u2581; unrelated symptoms." }, { "left": "issue:44060", "right": "issue:44077", "accept": false, - "reason": "Incorrect tied-weights warning and patchtsmixer post_init policy are unrelated." + "reason": "Incorrect tied-weights warning vs patchtsmixer post_init allowance; different configs." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "torch.compile attention-kernel divergence vs token roundtrip failure; no shared code path." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "Second-forward stale cache crash is unrelated to the flash_attention s_aux=None AttributeError." + "reason": "Second-forward stale cache crash vs flash_attention s_aux=None AttributeError." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed zero3 deque underflow vs stale auto_mappings reference to removed configs." }, { "left": "issue:43653", "right": "issue:44162", "accept": false, - "reason": "BigBirdTokenizer special-token registration bug is unrelated to ESM2 breakage." + "reason": "Tokenizer special-token decode bug vs ESM2 model breakage." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "A tie_word_embeddings warning is not the same as a Qwen3.5 flash-attention generation regression." - }, - { - "left": "issue:45137", - "right": "issue:45600", - "accept": false, - "reason": "ZeRO3 deque exhaustion is unrelated to stale auto_mappings references to removed Sam3LiteText configs." + "reason": "Tied-embeddings warning vs flash-attention generation regression; not the same failure." }, { "left": "issue:44877", "right": "issue:46170", "accept": false, - "reason": "Strict config loading for granite_speech is unrelated to ROCm/torch.distributed import failures." + "reason": "Config strictness issue vs Windows ROCm import/dependency issue." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "A tied-weights config warning is not the same as a stale num_return_sequences warning in paged generate()." - }, - { - "left": "issue:43673", - "right": "issue:44734", - "accept": false, - "reason": "Missing cache in GenerationMixin chunked_prefill and server KV-cache indexing crashes are different problems." + "reason": "Embedding warning vs stale num_return_sequences warning; different warnings and triggers." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and tie_word_embeddings warnings are unrelated." - }, - { - "left": "issue:44530", - "right": "issue:46170", - "accept": false, - "reason": "PagedAttentionCache group-type crash in Qwen3.5 is unrelated to Windows ROCm dependency issues." - }, - { - "left": "issue:43065", - "right": "issue:46123", - "accept": false, - "reason": "SAM3 pixel decoder placeholder conv and MaskGenerationPipeline final-batch handling are unrelated." - }, - { - "left": "issue:44933", - "right": "issue:45459", - "accept": false, - "reason": "Missing image_utils import and protobuf-related tokenizer error masking are unrelated." - }, - { - "left": "issue:44297", - "right": "issue:44623", - "accept": false, - "reason": "Tokenizer class mismatch on save_pretrained and processor save_pretrained missing files are different serialization bugs." + "reason": "Qwen3-VL weight-shape mismatch vs tied_word_embeddings warning; unrelated." } ] }, @@ -11653,11 +11521,10 @@ "nodes": [ "issue:28282", "issue:30990", - "issue:36010", "issue:41628", "issue:42371", - "issue:42673", "issue:42994", + "issue:43065", "issue:43232", "issue:43475", "issue:43482", @@ -11666,15 +11533,17 @@ "issue:43646", "issue:43673", "issue:43701", - "issue:43704", + "issue:43827", "issue:43873", "issue:44038", "issue:44162", "issue:44188", - "issue:44351", + "issue:44297", "issue:44361", + "issue:44530", "issue:44623", "issue:44661", + "issue:44734", "issue:44749", "issue:44869", "issue:44871", @@ -11684,34 +11553,39 @@ "issue:45092", "issue:45103", "issue:45230", - "issue:45278", "issue:45310", "issue:45356", "issue:45375", "issue:45381", "issue:45397", "issue:45406", + "issue:45459", "issue:45478", "issue:45656", - "issue:45753", "issue:46077", "issue:46123", "issue:46132", + "issue:46170", "issue:46202" ], "soft_pairs": [ - "issue:44869|issue:45356", + "issue:43673|issue:44734", + "issue:44530|issue:46170", + "issue:43065|issue:46123", + "issue:44933|issue:45459", + "issue:44297|issue:44623", "issue:44361|issue:44749", + "issue:44869|issue:45356", "issue:43646|issue:44162", "issue:43232|issue:43673", "issue:44623|issue:45656", "issue:44749|issue:45656", - "issue:43673|issue:45278", + "issue:43673|issue:43827", "issue:43701|issue:44749", "issue:44871|issue:46132", - "issue:42994|issue:43482", - "issue:43475|issue:45381", "issue:45230|issue:45397", + "issue:43475|issue:45381", + "issue:42994|issue:43482", "issue:43475|issue:43643", "issue:28282|issue:30990", "issue:44038|issue:44623", @@ -11725,236 +11599,229 @@ "issue:45081|issue:45103", "issue:45310|issue:45375", "issue:43873|issue:44945", - "issue:43232|issue:44188", - "issue:41628|issue:44351", - "issue:42673|issue:43704", - "issue:44945|issue:45753", - "issue:36010|issue:41628" + "issue:43232|issue:44188" ], - "prepared_review_unit_hash": "87711f993aae2a10ec2fe9a4f68ba4d685dcf803c3750fc6bb29dc2c24caffcf", + "prepared_review_unit_hash": "86466317d89bfac30aa9c0af7797f3e48649c91733898a8228a4beac43c768e3", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 35850, - "estimated_input_tokens": 8963, - "estimated_eval_tokens": 18182 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36879, + "estimated_input_tokens": 9220, + "estimated_eval_tokens": 18696 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous regression cluster with only a couple of true duplicates; the rest are related only at a broad Transformers-v5 / model-regression level, not the same underlying bug.", - "confidence": 0.87, - "canonical_issue_reason": "Issue 45278 is the best umbrella representative: it captures the broad upgrade-induced import/API breakage theme that recurs across the cluster and is more general than the niche model-specific reports.", + "summary": "This cluster is heterogeneous: most items are unrelated bugs that only share broad transformer subdomains (loading, generation, tokenizers, serving, quantization). The soft edges look like thematic similarity, not true duplicate pairs.", + "confidence": 0.9, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 45278 is the strongest global representative because it is the broadest, most cluster-like regression report rather than a narrow symptom or exact duplicate.", + "best_issue_reason": "issue:43873 is the broadest, most active report in the set, but the cluster is too mixed for it to serve as a true duplicate canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44869", - "right": "issue:45356", + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Different generation failures: missing cache state vs wrong tensor indexing in serve/KV continuation." + }, + { + "left": "issue:44530", + "right": "issue:46170", + "accept": false, + "reason": "PagedAttention crash on Qwen3.5 models is unrelated to a Windows/ROCm torch.distributed import dependency issue." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "SAM3 PixelDecoder dummy Conv2d is unrelated to MaskGenerationPipeline dropping a final partial batch." + }, + { + "left": "issue:44933", + "right": "issue:45459", + "accept": false, + "reason": "Missing import from image_utils is a different bug than protobuf-related tokenizer error masking." + }, + { + "left": "issue:44297", + "right": "issue:44623", "accept": false, - "reason": "Different failure modes: Whisper timestamp decoding crash vs Kimi tokenizer codec/regression issue." + "reason": "Both involve save_pretrained, but tokenizer metadata mismatch and processor missing files are different code paths." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "Tokenizer task AttributeError and a performance regression after upgrade are unrelated bugs." + "reason": "Tokenizer AttributeError on tasks is unrelated to the Chinese performance slowdown report." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper timestamp decode trailing-character crash is a different tokenizer regression than Kimi-K2.5 codec handling." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Custom AutoConfig missing fields is not the same as ESM2 being broken." + "reason": "AutoConfig trust_remote_code missing fields is unrelated to the ESM2 breakage report." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "Both are generation-related, but they target different code paths and symptoms." + "reason": "sync_gpus kwargs update bug and chunked_prefill cache-missing bug are distinct generation paths." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Processor save_pretrained missing files and double optimizer stepping in deepspeed are unrelated." + "reason": "processor.save_pretrained missing files is unrelated to deepspeed causing optimizer.step to run twice." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "A slowdown regression and a double-step optimizer bug do not match." + "reason": "Data filtering slowdown and double optimizer step are different failures." }, { "left": "issue:43673", - "right": "issue:45278", + "right": "issue:43827", "accept": false, - "reason": "Generation cache failure and broad import errors after upgrade are different issues." + "reason": "Generation cache regression is unrelated to docs still referencing removed pipeline APIs." }, { "left": "issue:43701", "right": "issue:44749", "accept": false, - "reason": "Checkpoint key mismatch is unrelated to the data-filtering slowdown report." + "reason": "resume_from_checkpoint key mismatch is unrelated to the filtering slowdown." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "EOS config mismatch is not the same as AttentionInterface.register changing registered behavior." + "reason": "Gemma eos_token_id mismatch is a config bug, not an AttentionInterface.register behavior change." }, { - "left": "issue:42994", - "right": "issue:43482", + "left": "issue:45230", + "right": "issue:45397", "accept": false, - "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are separate code paths." + "reason": "A generic bug report is not the same concrete issue as gemma-4 zero3 from_pretrained." }, { "left": "issue:43475", "right": "issue:45381", "accept": false, - "reason": "SAM 3 video encoder attribute error and Qwen2.5-VL position-id bug are different model-specific failures." + "reason": "SAM3 video encoder missing fpn_position_embeddings is unrelated to qwen2.5-vl video position_id logic." }, { - "left": "issue:45230", - "right": "issue:45397", + "left": "issue:42994", + "right": "issue:43482", "accept": false, - "reason": "A generic bug report does not establish the same Gemma-4 zero3 from_pretrained issue." + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different operations and failure modes." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "Different model families and different failure surfaces." + "reason": "SAM3 VisionEncoderOutput field omission is unrelated to AutoConfig trust_remote_code missing fields." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "Missing PyTorch import error is not the same as Sentence Transformers hanging on load." + "reason": "PyTorch missing ImportError and Sentence Transformers hanging on load are not the same bug." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe / transformers 5 bug and processor save_pretrained missing files are unrelated." + "reason": "Qwen3-VL-Moe breakage in transformers 5.0 is unrelated to processor.save_pretrained missing files." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Bidirectional attention mask construction and pipeline partial-batch handling are different bugs." + "reason": "Gemma4 attention-mask bug and MaskGenerationPipeline partial-batch bug are different code paths." }, { "left": "issue:45406", "right": "issue:46202", "accept": false, - "reason": "Gemma4Processor missing _tokenizer and Gemma pipeline roundtrip on \u2581 are not the same concrete failure." + "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Gemma pipeline roundtrip failure on \u2581." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 settings guidance and InternVL2 meta-init incompatibility are unrelated." + "reason": "TF32 API guidance is unrelated to InternVL2 meta-initialization incompatibility." }, { "left": "issue:41628", "right": "issue:44933", "accept": false, - "reason": "Both are import-related, but they concern different missing symbols and likely different breakages." + "reason": "Cannot import AutoImageProcessor from transformers is a different import issue than a nonexistent image_utils import." }, { "left": "issue:43475", "right": "issue:44661", "accept": false, - "reason": "SAM 3 video encoder error is unrelated to add-new-model-like / tokenizer mapping logic." + "reason": "SAM3 output attribute bug is unrelated to add-new-model-like failing inside TOKENIZER_MAPPING_NAMES." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized saving failure and missing pad_token_id on Llama4Config are different bugs." + "reason": "Quantized saving failure is unrelated to Llama4Config missing pad_token_id." }, { "left": "issue:45375", "right": "issue:45478", "accept": false, - "reason": "Missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig is a different issue from the from_pretrained error." + "reason": "Same model family, but config-field dropping and from_pretrained error are not the same concrete bug." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Mistral regex patch crash and auto_docstring annotations crash are unrelated implementation paths." + "reason": "Mistral regex patch crash and auto_docstring annotation crash are different AttributeErrors in different subsystems." }, { "left": "issue:45310", "right": "issue:45375", "accept": false, - "reason": "Both mention Qwen3.5 Moe, but one is a general from_pretrained error and the other is a missing vision-config field." + "reason": "Both mention Qwen3.5, but one is a from_pretrained error and the other is a missing config field being dropped." }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization offloading and pipeline-parallelism output corruption are separate problems." + "reason": "Quantization offloading issues are unrelated to incorrect output under pipeline parallelism." }, { "left": "issue:43232", "right": "issue:44188", "accept": false, - "reason": "Generation kwargs update after sync_gpus and attention-kernel divergence under torch.compile are different issues." - }, - { - "left": "issue:41628", - "right": "issue:44351", - "accept": false, - "reason": "AutoImageProcessor import failure and HybridCache import failure are distinct missing exports." - }, - { - "left": "issue:42673", - "right": "issue:43704", - "accept": true, - "reason": "Exact same title and same reported Qwen3ForCausalLM VRAM leak under multiple dataloader threads." - }, - { - "left": "issue:44945", - "right": "issue:45753", - "accept": false, - "reason": "Incorrect output under pipeline parallelism is not the same as an infinite loop on a specific image." - }, - { - "left": "issue:36010", - "right": "issue:41628", - "accept": false, - "reason": "GenerationMixin import failure and AutoImageProcessor import failure are unrelated import regressions." - }, - { - "left": "issue:45310", - "right": "issue:45478", - "accept": true, - "reason": "Same Qwen3.5 Moe from_pretrained regression; titles differ only in phrasing/version framing." + "reason": "Generation kwargs sync bug is unrelated to attention-kernel divergence under torch.compile." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "Grounded overall. The umbrella choice of issue 45278 is supported by the packet as a broad post-upgrade import-error report, and the two accepted soft pairs are exact/near-exact duplicates with matching titles and symptoms. The remaining soft-edge rejections are conservative and do not overclaim shared root cause." - }, - "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 }, @@ -11975,8 +11842,11 @@ "issue:30990", "issue:33290", "issue:35141", + "issue:36010", + "issue:41628", "issue:42175", "issue:42222", + "issue:42673", "issue:42831", "issue:42898", "issue:42994", @@ -11987,18 +11857,17 @@ "issue:43504", "issue:43526", "issue:43673", + "issue:43704", "issue:43723", "issue:43784", "issue:43824", - "issue:43874", "issue:43976", "issue:44038", - "issue:44060", "issue:44079", "issue:44188", "issue:44222", "issue:44242", - "issue:44448", + "issue:44351", "issue:44496", "issue:44568", "issue:44623", @@ -12007,39 +11876,41 @@ "issue:44792", "issue:44869", "issue:44933", + "issue:44945", "issue:44964", - "issue:45092", "issue:45103", - "issue:45125", "issue:45137", "issue:45200", "issue:45310", - "issue:45356", - "issue:45405", "issue:45561", "issue:45712", + "issue:45753", "issue:45812" ], "soft_pairs": [ + "issue:41628|issue:44351", + "issue:42673|issue:43704", + "issue:44945|issue:45753", "issue:44964|issue:45137", + "issue:36010|issue:41628", "issue:42831|issue:44222", - "issue:44222|issue:45310", "issue:42994|issue:43824", - "issue:43232|issue:44792", + "issue:44222|issue:45310", "issue:43673|issue:44188", + "issue:43232|issue:44792", "issue:44038|issue:45310", "issue:42175|issue:45712", "issue:43526|issue:45561", "issue:44496|issue:45561", "issue:43504|issue:45561", - "issue:43377|issue:44242", "issue:42175|issue:44933", + "issue:43377|issue:44242", "issue:42898|issue:44568", - "issue:42222|issue:45200", "issue:43976|issue:44933", + "issue:42222|issue:45200", "issue:30990|issue:35141", - "issue:43479|issue:44869", "issue:43278|issue:45137", + "issue:43479|issue:44869", "issue:43479|issue:44079", "issue:44623|issue:45310", "issue:30333|issue:33290", @@ -12047,240 +11918,232 @@ "issue:43784|issue:43824", "issue:44734|issue:45103", "issue:43723|issue:44568", - "issue:42898|issue:44779", - "issue:42898|issue:45356", - "issue:43874|issue:45092", - "issue:45405|issue:45712", - "issue:44060|issue:45125", - "issue:42898|issue:44448" + "issue:42898|issue:44779" ], - "prepared_review_unit_hash": "2dab2ddf9470aca44641fafeb520b35f44b08566a5ad5b7d2271744bbf3857e9", + "prepared_review_unit_hash": "ebb691b9a5cba425c7a93aa3884c8a71b4295484a152a3322428e0a653fa42de", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 37387, - "estimated_input_tokens": 9347, - "estimated_eval_tokens": 18950 + "soft_pair_count": 31, + "serialized_chars": 36849, + "estimated_input_tokens": 9213, + "estimated_eval_tokens": 18682 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "No soft edge looks like a true duplicate; the set is a heterogeneous mix of unrelated bugs with only superficial overlap in model families, tokenizer regressions, or shared subsystems.", - "confidence": 0.94, - "canonical_issue_reason": "42898 is the closest umbrella-style report: it describes a concrete v5 tokenizer regression with a clear before/after boundary, making it the best representative of the tokenizer-regression subset even though the overall cluster is not a single bug.", + "summary": "This cluster is mostly a grab-bag of distinct Transformers regressions (imports, tokenizer behavior, model loading/saving, multimodal bugs, training/runtime issues). Only one soft edge looks like a true duplicate: the two Qwen3ForCausalLM VRAM-leak reports.", + "confidence": 0.87, + "canonical_issue_reason": "issue:42673 is the clearest canonical issue because it is the fuller, earlier report of the same Qwen3ForCausalLM multi-dataloader-thread VRAM leak that 43704 appears to duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "42898 is the strongest stand-alone issue here because it is specific, user-facing, and framed as a regression with a reproducible version boundary; the other issues are either narrower manifestations or unrelated problems.", + "best_issue_reason": "issue:42673 is the best issue anchor in the cluster: it has the strongest duplicate evidence, a detailed repro, and the paired report 43704 looks like the same bug.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols (`AutoImageProcessor` vs `HybridCache`) and different code paths." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same title, same model, same symptom; this looks like the same Qwen3ForCausalLM VRAM-leak bug reported twice." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Different failures: incorrect LLM output under pipeline parallelism vs an infinite loop on a specific image for Qwen3_5." + }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Different problems: Phi-4 multimodal model loading vs DeepSpeed ZeRO3 deque underflow." + "reason": "Unrelated issues: Phi-4 multimodal model loading vs DeepSpeed ZeRO3 deque underflow." }, { - "left": "issue:42831", - "right": "issue:44222", + "left": "issue:36010", + "right": "issue:41628", "accept": false, - "reason": "Both mention FP8, but one is an accuracy regression and the other is a save_pretrained failure in MOE code paths." + "reason": "Both are import regressions, but they are different missing names and likely separate fixes." }, { - "left": "issue:44222", - "right": "issue:45310", + "left": "issue:42831", + "right": "issue:44222", "accept": false, - "reason": "Same model family, different failures: FP8 serialization vs from_pretrained import/loading error." + "reason": "Different bugs: FP8 accuracy drift vs FP8 `save_pretrained` failure in MoE." }, { "left": "issue:42994", "right": "issue:43824", "accept": false, - "reason": "Quantized model save failure is unrelated to an ImportError for Qwen2.5-VL class import." + "reason": "Quantized model saving and missing `Qwen2_5_VLForConditionalGeneration` import are separate problems." }, { - "left": "issue:43232", - "right": "issue:44792", + "left": "issue:44222", + "right": "issue:45310", "accept": false, - "reason": "Generation cache/sync_gpus bug vs janus image-generation test failure; no shared code-path." + "reason": "Both mention FP8/MoE, but one is save-time behavior and the other is a `from_pretrained` load error for a different Qwen model variant." }, { "left": "issue:43673", "right": "issue:44188", "accept": false, - "reason": "Chunked prefill cache issue is different from attention-kernel divergence under torch.compile branching." + "reason": "Different subsystems and symptoms: generation cache handling vs attention-kernel divergence under `torch.compile`." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Different concrete bugs: generation kwargs after `sync_gpus` vs a janus image-generation test failure." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Different Qwen variants and different symptoms: Qwen3-VL-Moe incompatibility vs Qwen3.5-Moe from_pretrained error." + "reason": "Both involve Qwen MoE models, but the model variants and failure modes differ; too broad to treat as the same bug." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "Backend dependency packaging issue vs leftover dummy class cleanup; not the same bug." + "reason": "Packaging/backend-dependency issue vs leftover dummy classes leaking into the module namespace." }, { "left": "issue:43526", "right": "issue:45561", "accept": false, - "reason": "ImageProcessorFast label reduction bug vs pytest-xdist file race in testing utils." + "reason": "Image-processor label reduction bug vs xdist race on a captured-info file are unrelated." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Unrecognized model/config loading failure vs test infrastructure race; unrelated." + "reason": "Model config loading failure vs test-worker file race; no shared code path." }, { "left": "issue:43504", "right": "issue:45561", "accept": false, - "reason": "Legacy preset loading issue vs captured_info.txt race condition; different code paths." + "reason": "Beit legacy-field load bug vs xdist test race are unrelated." }, { - "left": "issue:43377", - "right": "issue:44242", + "left": "issue:42175", + "right": "issue:44933", "accept": false, - "reason": "MIMI padding-mask inconsistency vs missing load-balancing loss when router logits are off." + "reason": "TensorFlow extra/backend packaging issue vs a missing `image_utils` import; different failure classes." }, { - "left": "issue:42175", - "right": "issue:44933", + "left": "issue:43377", + "right": "issue:44242", "accept": false, - "reason": "Packaging/backend selection issue vs a missing import symbol; no duplicate signal." + "reason": "MIMI padding-mask mismatch vs MoE load-balancing loss omission are different model bugs." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Both tokenizer-related, but one is clean_up_tokenization_spaces behavior and the other is add_special_tokens not adding BOS/EOS." + "reason": "Both are tokenizer regressions in v5, but one concerns cleanup spaces and the other special-token insertion for a specific tokenizer." }, { - "left": "issue:42222", - "right": "issue:45200", + "left": "issue:43976", + "right": "issue:44933", "accept": false, - "reason": "VitPose model breakage is unrelated to Gemma 4 token-type ID handling." + "reason": "Python version compatibility problem vs missing import from `image_utils`; not the same bug." }, { - "left": "issue:43976", - "right": "issue:44933", + "left": "issue:42222", + "right": "issue:45200", "accept": false, - "reason": "Python version compatibility issue vs missing import from image_utils; unrelated." + "reason": "Different models and symptoms: broken ViTPose models vs Gemma 4 missing default `mm_token_type_ids`." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "SentenceTransformers loading hang vs embedding reinitialization after resize; different failures." + "reason": "Sentence-Transformers loading hang and embedding reinitialization on resize are unrelated." }, { - "left": "issue:43479", - "right": "issue:44869", + "left": "issue:43278", + "right": "issue:45137", "accept": false, - "reason": "Config default initialization bug vs Whisper timestamp decode crash; not the same underlying defect." + "reason": "Embedding dtype drift during train/eval vs DeepSpeed ZeRO3 deque underflow are unrelated." }, { - "left": "issue:43278", - "right": "issue:45137", + "left": "issue:43479", + "right": "issue:44869", "accept": false, - "reason": "Dtype mismatch during evaluate vs DeepSpeed ZeRO3 deque error; different mechanisms." + "reason": "Phi-4 multimodal config defaults vs Whisper timestamp decoding crash are different code paths." }, { "left": "issue:43479", "right": "issue:44079", "accept": false, - "reason": "Default config initialization bug vs ModelOutput key assignment bug; separate APIs and symptoms." + "reason": "Config initialization with `None` and ModelOutput key assignment are separate abstractions and bugs." }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "Missing files on processor.save_pretrained is not the same as Qwen3.5-Moe from_pretrained failure." + "reason": "Processor save-pretrained file emission and Qwen3.5 Moe load errors are different operations and likely different fixes." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLflow reporting failure after training crash vs OOM with Adafactor in DeepSpeed; unrelated." + "reason": "MLflow reporting on failed training jobs vs Adafactor OOM in DeepSpeed are unrelated runtime issues." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one is clean-up behavior and the other is wrong token IDs for Granite; distinct bugs." + "reason": "Both are tokenizer regressions, but the affected models and concrete breakages differ." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Sentence-transformers import NameError vs missing Qwen2.5-VL export; unrelated import failures." + "reason": "Sentence-Transformers import failure due to `nn` vs missing export of a Qwen VL class are unrelated import problems." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "Serving KV-cache continuation crash vs auto_docstring annotations crash; different subsystems." + "reason": "Serving KV-cache indexing crash and auto-docstring annotation handling are unrelated." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Tokenizer loading failure vs special-token insertion regression; not the same code-path." + "reason": "Tokenizer loading failure in v5 vs special-token insertion regression for mdeberta-v3-base are different bugs." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different models and behaviors (cleanup spacing vs DeepSeek token IDs)." - }, - { - "left": "issue:42898", - "right": "issue:45356", - "accept": false, - "reason": "Kimi-K2.5 codec/regex regression is distinct from the general clean_up_tokenization_spaces behavior change." - }, - { - "left": "issue:43874", - "right": "issue:45092", - "accept": false, - "reason": "Missing image-patch helper in GLM46V is unrelated to InternVL2 meta-initialization incompatibility." - }, - { - "left": "issue:45405", - "right": "issue:45712", - "accept": false, - "reason": "PEFT version pinning problem vs dummy class cleanup; no duplicate relation." - }, - { - "left": "issue:44060", - "right": "issue:45125", - "accept": false, - "reason": "Incorrect tied-weights warning in Qwen3-Next is different from missing _tp_plan in Qwen3.5-Moe." - }, - { - "left": "issue:42898", - "right": "issue:44448", - "accept": false, - "reason": "Tokenizer cleanup regression vs Pegasus output mismatch across v4/v5; related theme but not the same bug." + "reason": "Both are tokenizer regressions, but one is a global v5 cleanup-space change and the other is a DeepSeek tokenizer correctness bug." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays mostly conservative. The one accepted soft edge (issue:42673 \u2194 issue:43704) is supported by identical titles and matching symptoms. The rejected soft edges are framed cautiously and do not overclaim duplicates. One minor wording point: the packet does not explicitly prove a \u2018detailed repro,\u2019 but the fuller body length and discussion activity make the anchor choice reasonable enough here." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -12300,14 +12163,13 @@ "issue:22355", "issue:28282", "issue:30333", - "issue:30990", "issue:34689", "issue:35141", "issue:36331", - "issue:38175", "issue:41628", "issue:42757", "issue:42831", + "issue:42898", "issue:43116", "issue:43122", "issue:43232", @@ -12315,29 +12177,30 @@ "issue:43644", "issue:43723", "issue:43761", + "issue:43874", + "issue:44060", "issue:44190", "issue:44206", "issue:44222", "issue:44263", "issue:44295", "issue:44297", - "issue:44360", "issue:44373", "issue:44448", "issue:44483", "issue:44484", "issue:44485", - "issue:44514", "issue:44623", - "issue:44871", "issue:44933", "issue:44977", "issue:45071", "issue:45092", "issue:45103", + "issue:45125", "issue:45245", "issue:45310", - "issue:45357", + "issue:45356", + "issue:45405", "issue:45406", "issue:45656", "issue:45712", @@ -12347,6 +12210,11 @@ "issue:46209" ], "soft_pairs": [ + "issue:42898|issue:45356", + "issue:43874|issue:45092", + "issue:45405|issue:45712", + "issue:44060|issue:45125", + "issue:42898|issue:44448", "issue:44295|issue:45656", "issue:44448|issue:46132", "issue:44933|issue:45712", @@ -12356,264 +12224,259 @@ "issue:43479|issue:43644", "issue:44190|issue:46082", "issue:44222|issue:44623", - "issue:42831|issue:44623", "issue:30333|issue:34689", + "issue:42831|issue:44623", "issue:43122|issue:46209", - "issue:45245|issue:46132", "issue:30333|issue:36331", "issue:22355|issue:28282", - "issue:45071|issue:46082", + "issue:45245|issue:46132", "issue:28282|issue:30333", - "issue:41628|issue:42757", "issue:30333|issue:35141", + "issue:45071|issue:46082", + "issue:41628|issue:42757", "issue:44206|issue:45103", "issue:44297|issue:45310", - "issue:44222|issue:44484", "issue:43723|issue:44977", "issue:43232|issue:45834", "issue:44263|issue:45092", "issue:43761|issue:46082", - "issue:44295|issue:44623", + "issue:44222|issue:44484", "issue:44373|issue:44485", - "issue:44483|issue:45406", - "issue:44514|issue:45103", - "issue:45092|issue:45357", - "issue:44263|issue:44360", - "issue:43723|issue:44871", - "issue:30990|issue:38175" + "issue:44295|issue:44623", + "issue:44483|issue:45406" ], - "prepared_review_unit_hash": "07fa8d08e8945a11893a5c630839b36ad36884cc0ece686ffd97048fea91fbc0", + "prepared_review_unit_hash": "54dc6f678cd45e18622c9327e613bfe8eace4f9bad4ac5d8fdc95cb649560203", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 34, - "serialized_chars": 37617, - "estimated_input_tokens": 9405, - "estimated_eval_tokens": 19066 + "serialized_chars": 37700, + "estimated_input_tokens": 9425, + "estimated_eval_tokens": 19106 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the soft pairs are mostly superficial text similarity around imports/loading/save_pretrained/v5 regressions, but they do not describe the same underlying bug or change. No soft edge should be merged.", - "confidence": 0.91, - "canonical_issue_reason": "No clear canonical issue: the items span unrelated bug classes and none of the soft pairs are duplicate-level matches.", + "summary": "These are mostly keyword-similar but substantively different issues; none of the soft pairs look like true duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue stands out: the set mixes unrelated import errors, model regressions, config validation bugs, and save/load edge cases.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good representative for the cluster; it is too mixed to treat any one issue as the canonical duplicate target.", + "best_issue_reason": "issue:45356 is the strongest standalone report here because it has a clear regression boundary, affected model, and concrete symptom, but it still should not absorb the rest of the cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they describe different failures: v5 clean_up_tokenization_spaces behavior vs Kimi-K2.5 codec handling and warning text." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Different multimodal bugs: one is a missing image-patch API, the other is a remote-code checkpoint/meta-init incompatibility." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "Unrelated: one is a dependency version bump problem, the other is cleanup of leftover dummy classes." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Both are Qwen-related, but one is a tied-weights warning and the other is missing tensor-parallel metadata; different code paths." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Both mention v5 regressions, but they affect different models and different behavior paths." + }, { "left": "issue:44295", "right": "issue:45656", "accept": false, - "reason": "Different subsystems and failure modes: buffer access around position_ids vs. double optimizer stepping with deepspeed." + "reason": "A buffer read error and a double optimizer step under DeepSpeed are unrelated bugs." }, { "left": "issue:44448", "right": "issue:46132", "accept": false, - "reason": "One is a model-output regression between v4 and v5; the other is altered behavior in AttentionInterface.register." + "reason": "Different problems: Pegasus output regression versus changed behavior in AttentionInterface.register." }, { "left": "issue:44933", "right": "issue:45712", "accept": false, - "reason": "Both concern imports/internal symbols, but one is a missing image_utils import and the other is leftover dummy PT classes leaking into dir()." + "reason": "Import/export failure in image_utils is unrelated to leftover dummy class cleanup." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "Both mention save_pretrained, but one is FP8 MOE serialization and the other is tokenizer metadata mismatch." + "reason": "Both involve save_pretrained, but one is FP8 MoE serialization and the other is tokenizer config metadata mismatch." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Unrelated: tokenizer save metadata bug vs. deepspeed optimizer stepping." + "reason": "Tokenizer save metadata and optimizer stepping are unrelated." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Different example scripts and failures: empty multilabel results vs. local dataset loading." + "reason": "Both are example-script reports, but one is multi-label classification output and the other is local dataset loading." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Different code paths: multimodal config defaults vs. non-persistent buffer initialization." + "reason": "Different subsystems: multimodal config defaults versus non-persistent buffer initialization in v5." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "Example-script dataset loading bug vs. LlamaConfig head_dim validation issue." + "reason": "Example dataset loading failure is unrelated to LlamaConfig head_dim validation." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "Both are serialization-related, but they affect different APIs and different missing/incorrect outputs." - }, - { - "left": "issue:42831", - "right": "issue:44623", - "accept": false, - "reason": "FP8 accuracy regression is unrelated to processor.save_pretrained missing files." + "reason": "Both touch save_pretrained, but one is model FP8/MoE behavior and the other is processor file emission." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "MLflow reporting on failed training is unrelated to a model-loading regression for Llama 3.2 Vision." + "reason": "MLflow job reporting on failure is unrelated to a model-loading regression." }, { - "left": "issue:43122", - "right": "issue:46209", + "left": "issue:42831", + "right": "issue:44623", "accept": false, - "reason": "Tokenization behavior change across versions is unrelated to FP8 inference failing on B200 GPUs." + "reason": "FP8 accuracy issues and missing processor save files are different bugs." }, { - "left": "issue:45245", - "right": "issue:46132", + "left": "issue:43122", + "right": "issue:46209", "accept": false, - "reason": "Category-count validation and AttentionInterface.register behavior are different bugs." + "reason": "Tokenizer differences across versions are unrelated to FP8 inference failures on B200." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "Failed-job MLflow reporting vs. Trainer compute_loss signature incompatibility are unrelated." + "reason": "A failed-job reporting bug and a Trainer compute_loss signature change are not the same issue." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "Both are import errors, but they concern different modules and different missing dependencies." + "reason": "Both are import errors, but one is a missing transformers.onnx module and the other is a missing PyTorch dependency." }, { - "left": "issue:45071", - "right": "issue:46082", + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category-count runtime error and AttentionInterface.register behavior change are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "Dependency import failure versus MLflow reporting bug; no shared underlying cause." + }, + { + "left": "issue:30333", + "right": "issue:35141", "accept": false, - "reason": "PretrainedConfig type-checking regression is unrelated to LlamaConfig head_dim rejection." + "reason": "Job-status reporting and embedding reinitialization are separate problems." }, { - "left": "issue:28282", - "right": "issue:30333", + "left": "issue:45071", + "right": "issue:46082", "accept": false, - "reason": "AutoModel missing torch and MLflow job-reporting are unrelated." + "reason": "Both are config-related, but one is type checking regression and the other is explicit head_dim validation." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Import failures from different packages and different causes; not the same bug." - }, - { - "left": "issue:30333", - "right": "issue:35141", - "accept": false, - "reason": "Training-job reporting issue vs. embedding reinitialization bug in post_init are unrelated." + "reason": "Different import/export issues in different packages, with different fixes." }, { "left": "issue:44206", "right": "issue:45103", "accept": false, - "reason": "Feature-extractor argument mismatch vs. auto-docstring crash under future annotations are different code paths." + "reason": "Unsupported processor argument and auto_docstring annotation crash are unrelated." }, { "left": "issue:44297", "right": "issue:45310", "accept": false, - "reason": "Tokenizer save metadata bug vs. Qwen3.5 MoE from_pretrained failure are distinct problems." - }, - { - "left": "issue:44222", - "right": "issue:44484", - "accept": false, - "reason": "FP8 MOE save_pretrained issue is unrelated to the max_shard_size default question." + "reason": "Tokenizer save metadata mismatch is unrelated to Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Tokenizer loading regression and Qwen3.5 flash-attention generation bug are different failures." + "reason": "Tokenizer loading and flash-attention generation issues affect different code paths." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation kwargs update bug is unrelated to Kosmos2.5 OCR index errors." + "reason": "Generation kwargs synchronization and OCR index overflow are unrelated." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "GlmMoeDsaIndexer torch.split issue is unrelated to InternVL2 meta-init compatibility." + "reason": "torch.split return handling in GlmMoeDsaIndexer is unrelated to meta-initialization incompatibility." }, { "left": "issue:43761", "right": "issue:46082", "accept": false, - "reason": "CLIP hidden_states regression and LlamaConfig head_dim validation are unrelated." + "reason": "CLIP hidden_states regression and LlamaConfig validation are different bugs." }, { - "left": "issue:44295", - "right": "issue:44623", + "left": "issue:44222", + "right": "issue:44484", "accept": false, - "reason": "Position_ids buffer access bug vs. processor.save_pretrained file omission are different issues." + "reason": "FP8 save_pretrained bug and a question about max_shard_size are not duplicates." }, { "left": "issue:44373", "right": "issue:44485", "accept": false, - "reason": "Docstring wording for position_ids is not the same as a RoPE implementation discussion/bug." - }, - { - "left": "issue:44483", - "right": "issue:45406", - "accept": false, - "reason": "Chat-completions request handling and Gemma4Processor _tokenizer crash are different serve-side problems." - }, - { - "left": "issue:44514", - "right": "issue:45103", - "accept": false, - "reason": "Qwen2_5_VL batched chat-template crash is unrelated to the auto-docstring annotations bug." - }, - { - "left": "issue:45092", - "right": "issue:45357", - "accept": false, - "reason": "Old InternVL2 remote-code meta-init incompatibility vs. Qwen3.5 visual encoder key saving regression are different bugs." - }, - { - "left": "issue:44263", - "right": "issue:44360", - "accept": false, - "reason": "Both mention DSA indexer, but one is a torch.split value bug and the other is about a missing ReLU; not the same underlying defect." + "reason": "Wrong docstring and RoPE implementation discussion do not describe the same issue." }, { - "left": "issue:43723", - "right": "issue:44871", + "left": "issue:44295", + "right": "issue:44623", "accept": false, - "reason": "Tokenizer loading regression and Gemma-3 eos_token_id inconsistency are separate configuration/loading issues." + "reason": "Reading a registered buffer and missing processor save files are unrelated." }, { - "left": "issue:30990", - "right": "issue:38175", + "left": "issue:44483", + "right": "issue:45406", "accept": false, - "reason": "Sentence-Transformers loading hang and SigLIP2 zero-probabilities behavior are unrelated." + "reason": "Chat-completions request handling and Gemma4Processor missing _tokenizer are different failures." } ] }, @@ -12640,7 +12503,7 @@ "issue:22355", "issue:28218", "issue:30990", - "issue:33357", + "issue:38175", "issue:39401", "issue:41628", "issue:42371", @@ -12650,7 +12513,6 @@ "issue:42907", "issue:42994", "issue:43066", - "issue:43122", "issue:43232", "issue:43479", "issue:43576", @@ -12662,24 +12524,24 @@ "issue:44038", "issue:44060", "issue:44117", - "issue:44188", "issue:44261", "issue:44263", - "issue:44295", - "issue:44297", + "issue:44360", "issue:44373", "issue:44442", - "issue:44521", + "issue:44514", "issue:44568", "issue:44671", "issue:44704", "issue:44829", + "issue:44871", "issue:44913", "issue:44991", "issue:45092", + "issue:45103", "issue:45278", - "issue:45290", "issue:45310", + "issue:45357", "issue:45593", "issue:45753", "issue:46002", @@ -12687,9 +12549,14 @@ "issue:46132" ], "soft_pairs": [ + "issue:44263|issue:44360", + "issue:44514|issue:45103", + "issue:45092|issue:45357", + "issue:43723|issue:44871", + "issue:30990|issue:38175", "issue:42907|issue:46050", - "issue:43723|issue:44991", "issue:44261|issue:45092", + "issue:43723|issue:44991", "issue:44829|issue:46002", "issue:43723|issue:44442", "issue:41628|issue:43824", @@ -12705,213 +12572,208 @@ "issue:22355|issue:28218", "issue:42831|issue:45753", "issue:43756|issue:44060", + "issue:44117|issue:44704", "issue:44671|issue:46132", "issue:43576|issue:43784", - "issue:44117|issue:44704", "issue:43646|issue:46002", - "issue:30990|issue:33357", - "issue:43122|issue:44521", - "issue:42371|issue:44263", - "issue:44295|issue:44297", - "issue:44704|issue:45290", - "issue:44188|issue:44373" + "issue:42371|issue:44263" ], - "prepared_review_unit_hash": "4a0350ebd7d4be3093f3494747c3e1b9597a5cb77a4daca09f44b8d65b41fe59", + "prepared_review_unit_hash": "501ad87e7f6f3f9299b96610247ce7c58513db3d00f5bfe33bb485cacbc2eb41", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35686, - "estimated_input_tokens": 8922, - "estimated_eval_tokens": 18100 + "serialized_chars": 35135, + "estimated_input_tokens": 8784, + "estimated_eval_tokens": 17824 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose grab-bag of Transformers regressions, mostly around v5 tokenizer/import/config/model-init behavior. The items are thematically related but do not look like one duplicate bug cluster; I found no soft pair that clearly shares the same concrete root cause.", + "summary": "The cluster is mostly a grab-bag of unrelated Transformers regressions; the soft-similarity pairs share broad themes (tokenizers, import errors, save/load, multimodal configs) but not the same concrete bug. No soft edge is safe to merge.", "confidence": 0.87, - "canonical_issue_reason": "issue:43723 is the broadest, most canonical tokenizer-loading regression here; it names the shared from_pretrained path that several other tokenizer-related reports are adjacent to, even though the others are not proven duplicates.", + "canonical_issue_reason": "issue:43723 is the best umbrella representative because it is a broad, user-facing tokenizer-loading regression in v5 and is closest to several other tokenizer/import complaints without being overly narrow.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43723 is the best single issue to anchor this set because it is general, user-facing, and closest to the recurring tokenizer-load failures in the group.", + "best_issue_reason": "issue:43723 is the strongest global representative for the cluster: it captures the recurring tokenizer/from_pretrained failure mode that appears most central here, while the other issues are mostly narrower or unrelated variants.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:42907", - "right": "issue:46050", + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Different DSA-related bugs: one is about torch.split return values, the other about a missing ReLU. Same subsystem, not the same defect." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "Completely different code paths: Qwen2.5-VL chat template batching versus auto_docstring annotation handling." + }, + { + "left": "issue:45092", + "right": "issue:45357", "accept": false, - "reason": "Both concern quantized save/load, but they describe different quantization workflows and failure modes, so they do not look like the same bug." + "reason": "Both are multimodal model issues, but one is meta-init incompatibility for old remote-code checkpoints and the other is incorrect visual encoder keys on save_pretrained." }, { "left": "issue:43723", - "right": "issue:44991", + "right": "issue:44871", + "accept": false, + "reason": "Tokenizer-loading regression versus an eos_token_id configuration mismatch; related area, but different bug." + }, + { + "left": "issue:30990", + "right": "issue:38175", "accept": false, - "reason": "Both are tokenizer loading failures in v5, but one is generic and the other is model-specific; no clear evidence they share the same root cause." + "reason": "Different models and symptoms: loading gets stuck versus zero probabilities in SigLIP2." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both involve quantized model persistence, but they target different quantization flows and failure modes." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "These are different problems: a missing rms_norm_eps in MLA q_a_layernorm versus remote-code/meta-initialization incompatibility." + "reason": "Layernorm precision/config issue is unrelated to meta-initialization compatibility for remote-code checkpoints." + }, + { + "left": "issue:43723", + "right": "issue:44991", + "accept": false, + "reason": "Both mention tokenizer loading, but the titles point to different models and likely different root causes." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "Different symptoms and code paths: flash_attention_3 training degeneration versus NaNs during checkpoint/init of classification heads." + "reason": "Flash-attention training degradation and NaNs in classification heads are different failures, not one bug." }, { "left": "issue:43723", "right": "issue:44442", "accept": false, - "reason": "Both involve AutoTokenizer, but one is a general v5 loading regression and the other is a specific tokenizer class mapping failure." + "reason": "AutoTokenizer failing on one tokenizer class is not the same bug as the broader v5 tokenizer-loading regression." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Both are import errors, but they target different symbols and likely different missing-export paths." + "reason": "Both are import errors, but for different symbols and likely different export/mapping causes." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "An accuracy regression in FineGrainedFP8 is unrelated to a Qwen3.5 MoE from_pretrained error." + "reason": "FineGrainedFP8 accuracy issue is unrelated to Qwen3.5 MoE from_pretrained failure." }, { "left": "issue:43479", "right": "issue:44913", "accept": false, - "reason": "Different config bugs: multimodal sub-config defaults versus rotary_pct not persisting on reload." + "reason": "Different configuration bugs in different model families: default multimodal configs versus rotary_pct reload behavior." }, { "left": "issue:44263", "right": "issue:44373", "accept": false, - "reason": "A torch.split indexing issue is unrelated to a docstring typo for position_ids." + "reason": "One concerns torch.split return values, the other a docstring error; not the same bug." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "These are unrelated model/quantization issues with no shared concrete code path." + "reason": "Different Qwen/FP8 versus Qwen3-VL-Moe issues with no concrete shared code-path." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Wrong offset mappings and missing BOS/EOS behavior are distinct tokenizer regressions." + "reason": "Offset_mapping behavior and BOS/EOS special-token insertion are distinct tokenizer bugs." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "A specific huggingface_hub import failure is not the same as a broad set of import errors after upgrading to v5.5.0." + "reason": "Missing is_offline_mode import is not the same as a broad set of import errors after upgrading to v5.5.0." }, { "left": "issue:43232", "right": "issue:45593", "accept": false, - "reason": "Generation kwargs update behavior and missing auxiliary losses in D-FINE are unrelated." + "reason": "Generation kwargs sync issue and D-FINE auxiliary-loss behavior are unrelated." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Offline tokenizer cache loading and quantized model saving are different failure domains." + "reason": "Offline tokenizer cache loading and quantized model saving are different persistence problems." }, { "left": "issue:42886", "right": "issue:43066", "accept": false, - "reason": "Both touch tokenizer handling, but offline cache loading is not the same bug as an incorrect tokenizer decoder type in v5." + "reason": "Both concern tokenizers, but one is offline cache loading and the other is decoder-type configuration in v5." }, { "left": "issue:22355", "right": "issue:28218", "accept": false, - "reason": "A missing transformers.onnx module and an added-token spacing bug are unrelated." + "reason": "No module import error for transformers.onnx is unrelated to added-token spacing behavior." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "FineGrainedFP8 accuracy regression and a Qwen3_5 infinite loop are clearly different issues." + "reason": "FP8 accuracy regression and an infinite-generation loop on a specific image are unrelated." }, { "left": "issue:43756", "right": "issue:44060", "accept": false, - "reason": "Smollm3 RoPE-layer loss and Qwen3-Next tied-weights warnings are different model-specific bugs." + "reason": "RoPE-layer omission and tied-weights warning are different model-architecture bugs." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES None handling and AutoProcessor kwargs propagation are separate from_pretrained issues." }, { "left": "issue:44671", "right": "issue:46132", "accept": false, - "reason": "CamemBERT masked-LM mispredictions and AttentionInterface.register behavior changes do not share the same code path." + "reason": "CamemBERT prediction regression is unrelated to AttentionInterface.register behavior changes." }, { "left": "issue:43576", "right": "issue:43784", "accept": false, - "reason": "A broken CLI command and an import-time NameError in sentence-transformers are unrelated." - }, - { - "left": "issue:44117", - "right": "issue:44704", - "accept": false, - "reason": "A tokenizer mapping None bug and an AutoProcessor kwargs passthrough issue are different dispatcher/cached_file problems." + "reason": "Broken env command and sentence-transformers import NameError are different regressions." }, { "left": "issue:43646", "right": "issue:46002", "accept": false, - "reason": "Custom model initialization breakage and NaNs in classification heads are not the same underlying bug." - }, - { - "left": "issue:30990", - "right": "issue:33357", - "accept": false, - "reason": "Sentence-Transformers loading stall and a macOS bus error on CLIP are distinct runtime failures." - }, - { - "left": "issue:43122", - "right": "issue:44521", - "accept": false, - "reason": "Tokenization changes across versions and multimodal assistant mask generation are separate behaviors." + "reason": "Custom model initialization breakage is not the same as NaNs in classification heads." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 API guidance and torch.split return-value handling are unrelated." - }, - { - "left": "issue:44295", - "right": "issue:44297", - "accept": false, - "reason": "Reading position_ids after buffer registration and tokenizer_class mismatch on save_pretrained are different serialization issues." - }, - { - "left": "issue:44704", - "right": "issue:45290", - "accept": false, - "reason": "Passing kwargs to cached_file and crashing on tool-call assistant messages are unrelated apply_chat_template/processor paths." - }, - { - "left": "issue:44188", - "right": "issue:44373", - "accept": false, - "reason": "torch.compile attention kernel divergence is unrelated to a docstring error." + "reason": "TF32 settings misuse is unrelated to the DSA indexer split/ReLU bugs." } ] }, @@ -12939,16 +12801,15 @@ "issue:28218", "issue:29159", "issue:30990", + "issue:33357", "issue:36296", "issue:39401", "issue:41628", - "issue:42175", "issue:42757", "issue:42907", "issue:43116", "issue:43122", "issue:43232", - "issue:43278", "issue:43408", "issue:43452", "issue:43475", @@ -12958,6 +12819,7 @@ "issue:43937", "issue:43976", "issue:44038", + "issue:44188", "issue:44222", "issue:44261", "issue:44295", @@ -12970,259 +12832,275 @@ "issue:44514", "issue:44521", "issue:44625", + "issue:44704", "issue:44743", "issue:44779", "issue:44829", - "issue:44938", "issue:44945", "issue:44964", "issue:45200", "issue:45216", "issue:45278", + "issue:45290", "issue:45446", "issue:45478", "issue:45600" ], "soft_pairs": [ - "issue:42907|issue:44945", + "issue:43122|issue:44521", + "issue:30990|issue:33357", + "issue:44295|issue:44297", + "issue:44188|issue:44373", + "issue:44704|issue:45290", "issue:41628|issue:45278", + "issue:42907|issue:44945", + "issue:44261|issue:44373", "issue:43232|issue:44743", "issue:43784|issue:45600", - "issue:44261|issue:44373", - "issue:43976|issue:44945", "issue:39401|issue:44488", + "issue:43976|issue:44945", "issue:43408|issue:43475", - "issue:43122|issue:44514", "issue:42907|issue:43937", + "issue:43122|issue:44514", "issue:39401|issue:44521", - "issue:44625|issue:44829", "issue:44360|issue:44485", + "issue:44625|issue:44829", "issue:42907|issue:44964", "issue:43116|issue:45200", "issue:44038|issue:44297", - "issue:28218|issue:30990", "issue:44038|issue:44295", + "issue:28218|issue:30990", "issue:43122|issue:44779", "issue:42907|issue:43452", "issue:36296|issue:43116", "issue:43824|issue:44351", "issue:44038|issue:44222", - "issue:43122|issue:43906", "issue:45216|issue:45478", - "issue:42757|issue:44351", + "issue:43122|issue:43906", "issue:43784|issue:45446", - "issue:43278|issue:44945", - "issue:22355|issue:29159", - "issue:42175|issue:44938" + "issue:42757|issue:44351", + "issue:22355|issue:29159" ], - "prepared_review_unit_hash": "3a96a2be2c6d5d89cd5f53d746ed0c0441b0a2ad8d997ed1511c3039a344e172", + "prepared_review_unit_hash": "6049652fbcc15fc4679699e7d40e2dbcd64c14d11ff0d5fad82717484e954248", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 35658, - "estimated_input_tokens": 8915, - "estimated_eval_tokens": 18086 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37204, + "estimated_input_tokens": 9301, + "estimated_eval_tokens": 18858 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The proposed soft links are all false positives: they span unrelated tokenizer, import, model-loading, configuration, and training bugs rather than one shared defect. I picked a tokenizer-regression issue as the best available representative, but the cluster is not a true duplicate set.", - "confidence": 0.84, - "canonical_issue_reason": "issue:43122 is the broadest and most central tokenizer-regression report in this set, with multiple tokenizer-related soft links nearby.", + "summary": "This cluster is highly heterogeneous: most items are unrelated regressions or import/model issues that only share a broad subsystem. I would anchor on issue:43122 as the most representative tokenizer-regression report, but I do not see any soft pair that is clearly the same concrete bug/change.", + "confidence": 0.68, + "canonical_issue_reason": "issue:43122 is the broadest and most central tokenizer-regression report here, and it overlaps most with the other tokenizer-related items in the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43122 is the best cluster representative because it is a general regression description and sits closest to the tokenizer-related subset of the noisy cluster.", + "best_issue_reason": "It is the best representative issue for the tokenizer-regression slice of this mixed cluster: specific enough to be actionable, but broad enough to cover the nearby tokenization complaints.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:42907", - "right": "issue:44945", + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization/chat-template behavior, but one is a generic tokenizer regression and the other is multimodal assistant_mask generation; different defect." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Different failures: Sentence-Transformers loading hang vs a Mac bus error in CLIP model loading." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Unrelated symptoms: position_ids buffer access error vs tokenizer_class mismatch in tokenizer_config.json." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to a wrong docstring for position_ids." + }, + { + "left": "issue:44704", + "right": "issue:45290", "accept": false, - "reason": "Dequantized save_pretrained failure and pipeline-parallel generation corruption are different code paths and different bugs." + "reason": "Different code paths: AutoProcessor kwargs/cached_file plumbing vs apply_chat_template crashing on tool-call messages." }, { "left": "issue:41628", "right": "issue:45278", "accept": false, - "reason": "Missing AutoImageProcessor export and broad import failures after upgrade are related only at a high level, not the same defect." + "reason": "Both are import-related, but one is a missing AutoImageProcessor export and the other is a broader upgrade breakage; not the same bug." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Saving dequantized models and incorrect pipeline-parallel outputs are different problems." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "Precision/config mismatch in MLA layernorm vs a docstring issue; no duplicate signal." }, { "left": "issue:43232", "right": "issue:44743", "accept": false, - "reason": "Generation kwargs after sync_gpus and Qwen3 recurrent-state reset are unrelated generation issues." + "reason": "Both touch generation internals, but sync_gpus kwarg handling and Qwen3.5 recurrent-state resets are distinct defects." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "Sentence-transformers import NameError and auto_mappings CI break from removed configs are different failure modes." + "reason": "Import NameError in sentence-transformers integration vs stale auto_mappings references; different breakages." }, { - "left": "issue:44261", - "right": "issue:44373", + "left": "issue:39401", + "right": "issue:44488", "accept": false, - "reason": "A precision/config bug in MLA q_a_layernorm is not the same as a docstring typo for position_ids." + "reason": "Qwen3 offset_mapping is unrelated to a model-loading failure for cjvt/sleng-bert." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Python version compatibility and incorrect pipeline-parallel output are different problems." - }, - { - "left": "issue:39401", - "right": "issue:44488", - "accept": false, - "reason": "Wrong offset mappings for Qwen3 tokenizer and failure to load cjvt/sleng-bert are unrelated." + "reason": "Python version support regression is unrelated to incorrect outputs under pipeline parallelism." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "Model-type warning for sam3_video vs sam3_tracker is not the same as a missing Sam3VisionEncoderOutput attribute." + "reason": "Both are SAM3-related, but a model-type warning and a missing attribute on vision encoder output are not the same bug." }, { - "left": "issue:43122", - "right": "issue:44514", + "left": "issue:42907", + "right": "issue:43937", "accept": false, - "reason": "Tokenizer behavior changed across versions and batched apply_chat_template crashing are different concrete bugs." + "reason": "Dequantized save failure vs invalid GenerationConfig for GLM-5 are different issues." }, { - "left": "issue:42907", - "right": "issue:43937", + "left": "issue:43122", + "right": "issue:44514", "accept": false, - "reason": "Saving dequantized models and invalid GenerationConfig for GLM-5 do not share the same code path." + "reason": "Tokenizer version-drift regression vs batched apply_chat_template crash with padding=False; different concrete symptoms." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Wrong offset_mapping and zero assistant_masks are both tokenizer-related, but they are distinct outputs and root causes." + "reason": "Different multimodal/tokenizer bugs: wrong offset_mapping vs assistant_masks all zero." }, { - "left": "issue:44625", - "right": "issue:44829", + "left": "issue:44360", + "right": "issue:44485", "accept": false, - "reason": "num_labels propagation in Qwen3.5 and flash_attention_3 degenerate training are unrelated." + "reason": "DSA indexer/ReLU discussion is unrelated to GLM-5 RoPE implementation." }, { - "left": "issue:44360", - "right": "issue:44485", + "left": "issue:44625", + "right": "issue:44829", "accept": false, - "reason": "DSA indexer ReLU and GLM-5 RoPE implementation are different subsystems with no concrete overlap." + "reason": "Qwen3.5 num_labels propagation and flash_attention_3 training degeneration are different code paths." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Dequantized save_pretrained failure and loading Phi-4 multimodal are separate model I/O issues." + "reason": "Saving dequantized checkpoints is unrelated to a model-load failure for Phi-4 multimodal instruct." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Empty multi-label classification results and missing mm_token_type_ids default in Gemma 4 are not the same bug." + "reason": "Example-script multi-label empties and Gemma 4 mm_token_type_ids defaulting are different bugs." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe load bug and tokenizer_class mismatch on save_pretrained affect different flows." + "reason": "Qwen3-VL-Moe issue vs tokenizer_class mismatch; no shared concrete defect." }, { - "left": "issue:28218", - "right": "issue:30990", + "left": "issue:44038", + "right": "issue:44295", "accept": false, - "reason": "Extra space after added token and Sentence Transformers getting stuck loading are unrelated." + "reason": "Qwen3-VL-Moe bug is unrelated to the position_ids buffer access problem." }, { - "left": "issue:44038", - "right": "issue:44295", + "left": "issue:28218", + "right": "issue:30990", "accept": false, - "reason": "Qwen3-VL-Moe failure and position_ids buffer read error are different code paths." + "reason": "Tokenizer spacing bug vs Sentence-Transformers loading hang; unrelated." }, { "left": "issue:43122", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer regressions, but they are model-specific and do not point to the same underlying fix." + "reason": "Both are tokenizer regressions in v5, but one is broad version drift and the other is Deepseek-specific; not enough evidence of the same root cause." }, { "left": "issue:42907", "right": "issue:43452", "accept": false, - "reason": "Dequantized save failure and gguf_file breaking from_pretrained are different loading/saving bugs." + "reason": "Dequantized save_pretrained failure vs gguf_file loading breakage are different artifacts." }, { "left": "issue:36296", "right": "issue:43116", "accept": false, - "reason": "Tensor parallel training bug and multi-label classification script returning empty results are unrelated." + "reason": "Tensor-parallel training bug and example-script classification output bug are unrelated." }, { "left": "issue:43824", "right": "issue:44351", "accept": false, - "reason": "Different missing symbols: Qwen2_5_VLForConditionalGeneration vs HybridCache; both are import errors but not the same one." + "reason": "Different missing symbols: Qwen2_5_VLForConditionalGeneration vs HybridCache." }, { "left": "issue:44038", "right": "issue:44222", "accept": false, - "reason": "Qwen3-VL-Moe load bug and FP8 save_pretrained moe concern different functionality." - }, - { - "left": "issue:43122", - "right": "issue:43906", - "accept": false, - "reason": "Tokenizer regression and isolated reproduction of another issue are too vague to treat as the same concrete bug." - }, - { - "left": "issue:45216", - "right": "issue:45478", - "accept": false, - "reason": "Qwen3.5 save_pretrained regression and Qwen3.5 Moe from_pretrained error are different operations and likely different causes." + "reason": "Both mention MoE, but one is a model-specific loading/behavior issue and the other is FP8 save_pretrained; different fixes." }, { - "left": "issue:42757", - "right": "issue:44351", + "left": "issue:45216", + "right": "issue:45478", "accept": false, - "reason": "Import errors for is_offline_mode and HybridCache are separate missing-symbol issues." + "reason": "Related Qwen3.5 family, but save_pretrained producing a bad checkpoint and from_pretrained throwing an error are not clearly the same concrete bug." }, { "left": "issue:43784", "right": "issue:45446", "accept": false, - "reason": "Import NameError in sentence-transformers and a PyTorch version check bug in flex_attention are unrelated." + "reason": "Importing sentence-transformers NameError and a PyTorch version check in flex_attention are unrelated." }, { - "left": "issue:43278", - "right": "issue:44945", + "left": "issue:42757", + "right": "issue:44351", "accept": false, - "reason": "BF16-to-FP32 embedding dtype drift during evaluation is not the same as wrong LLM output under pipeline parallelism." + "reason": "Missing is_offline_mode from huggingface_hub and missing HybridCache from transformers are different import failures." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "Missing transformers.onnx module and slow-vs-fast tokenizer inconsistency are unrelated." - }, - { - "left": "issue:42175", - "right": "issue:44938", - "accept": false, - "reason": "TensorFlow backend missing in a torch install and Python 3.14 load failure are different environment/package problems." + "reason": "No module named transformers.onnx is unrelated to slow-vs-fast tokenizer inconsistency." } ] }, @@ -13251,14 +13129,15 @@ "issue:33453", "issue:38617", "issue:39401", + "issue:42175", "issue:42222", "issue:42371", "issue:42673", "issue:43064", "issue:43066", "issue:43116", + "issue:43278", "issue:43441", - "issue:43450", "issue:43493", "issue:43611", "issue:43673", @@ -13268,25 +13147,23 @@ "issue:43906", "issue:43937", "issue:43976", - "issue:43986", "issue:43992", "issue:44038", "issue:44075", "issue:44117", "issue:44291", "issue:44380", - "issue:44466", "issue:44485", "issue:44704", "issue:44779", "issue:44936", - "issue:44977", + "issue:44938", + "issue:44945", "issue:45071", "issue:45072", "issue:45083", "issue:45278", "issue:45357", - "issue:45405", "issue:45446", "issue:45478", "issue:45563", @@ -13296,245 +13173,231 @@ "issue:46077" ], "soft_pairs": [ - "issue:43493|issue:45753", + "issue:43278|issue:44945", + "issue:42175|issue:44938", "issue:42673|issue:45072", + "issue:43493|issue:45753", "issue:38617|issue:45278", - "issue:43992|issue:44704", "issue:43066|issue:44779", + "issue:43992|issue:44704", "issue:22355|issue:29127", "issue:43611|issue:46002", "issue:43937|issue:44485", "issue:43976|issue:45446", "issue:42222|issue:43441", - "issue:43116|issue:43493", "issue:42371|issue:43874", + "issue:43116|issue:43493", "issue:44075|issue:44936", - "issue:43064|issue:43673", "issue:45357|issue:45478", "issue:44038|issue:45478", + "issue:43064|issue:43673", "issue:43116|issue:43825", "issue:45083|issue:45446", "issue:43937|issue:45563", "issue:44380|issue:46077", - "issue:33453|issue:39401", "issue:44117|issue:45588", + "issue:33453|issue:39401", "issue:43827|issue:45446", "issue:39401|issue:44779", "issue:42673|issue:43906", - "issue:44291|issue:45071", - "issue:43937|issue:44977", - "issue:43450|issue:43986", - "issue:44466|issue:46002", - "issue:45405|issue:45446" + "issue:44291|issue:45071" ], - "prepared_review_unit_hash": "664bd30520d99a718a6c766c1df2cbd80089c05d8729e3f6c77c41dbeb72baa5", + "prepared_review_unit_hash": "f43a8972b49992175556c075559720b0bac1bbe746631f45409a3bb8848a309a", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 30, - "serialized_chars": 36436, - "estimated_input_tokens": 9109, - "estimated_eval_tokens": 18474 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35172, + "estimated_input_tokens": 8793, + "estimated_eval_tokens": 17842 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are mostly heterogeneous Transformers bug reports that only overlap at a broad theme level (v5 regressions, tokenizer/model issues, and dependency/version breakage). I don\u2019t see any soft pair that looks like the same underlying bug or change.", - "confidence": 0.83, - "canonical_issue_reason": "issue:45278 is the best umbrella candidate because its title is broad and migration-oriented (\u201cmany import errors after update\u201d), making it the closest representative of the cluster\u2019s recurring v5 breakage theme.", + "summary": "The cluster is thematically noisy but mostly centers on Transformers v5 regressions, tokenizer/loading issues, and model-specific bugs. The soft pairs do not look like true duplicates under a strict code-path standard, so all candidate edges are rejected.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45278 is the broadest umbrella report here: it captures widespread v5 import breakage after upgrade, making it the closest cluster anchor.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the most globally representative issue in this set: it is broad, user-facing, and plausibly subsumes several import/breakage reports better than the narrowly scoped model-specific tickets.", + "best_issue_reason": "issue:45278 is the best overall representative because it is the most general and high-impact regression report in the set, though it is not a safe duplicate sink for the more specific issues.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43493", - "right": "issue:45753", + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Different bugs: dtype drift between train/eval vs incorrect output under pipeline parallelism." + }, + { + "left": "issue:42175", + "right": "issue:44938", "accept": false, - "reason": "Same model family, but one is a SigLIP2 implementation discrepancy and the other is an infinite-loop generation bug on a specific image; different code paths and symptoms." + "reason": "Packaging/backend dependency issue vs Python 3.14 import/load failure." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM threading is unrelated to SGD argument handling / dtype mismatch in CI." + "reason": "VRAM leak in threaded dataloader use vs bfloat16 dtype mismatch in inference." }, { - "left": "issue:38617", - "right": "issue:45278", + "left": "issue:43493", + "right": "issue:45753", "accept": false, - "reason": "Both concern import failures after v5 changes, but one is a specific missing symbol and the other is a broad upgrade-wide report; not the same concrete breakage." + "reason": "SigLIP2 implementation discrepancy vs Qwen3_5 infinite loop on a specific image." }, { - "left": "issue:43992", - "right": "issue:44704", + "left": "issue:38617", + "right": "issue:45278", "accept": false, - "reason": "Missing `embed_tokens.weight` in UMT5Encoder is unrelated to `AutoProcessor.from_pretrained` not forwarding kwargs to `cached_file`." + "reason": "Both are import-related, but one is a specific missing symbol and the other is a broad upgrade report; not the same concrete failure." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer-related, but the reported defects differ: wrong decoder type vs incorrect tokenization results for Deepseek." + "reason": "Tokenizer decoder type regression vs Deepseek tokenizer output corruption; related area, distinct bug." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Missing model weight on load vs AutoProcessor not forwarding kwargs to cached_file." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "`transformers.onnx` missing module import and a layoutlmv3 error-message clarity issue are unrelated." + "reason": "Missing module import vs LayoutLMv3 error-message clarity; unrelated." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "`base_model_prefix` loading breakage and NaNs in classification heads are different failures with no shared code-path evidence." + "reason": "Model loading breakage from base_model_prefix vs NaNs in classification heads after init/checkpoint." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "GenerationConfig validation error and GLM-5 RoPE implementation concern different model-generation subsystems." + "reason": "GenerationConfig validation failure vs GLM-5 RoPE implementation discussion; not the same code path." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "These are both version-check bugs, but in different areas (`Python` support vs PyTorch import gating) and not the same underlying defect." + "reason": "Python support/version compatibility issue vs PyTorch version guard for AuxRequest import." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "VitPose model breakage and Ministral-3 FlashAttention failure are unrelated model-specific issues." - }, - { - "left": "issue:43116", - "right": "issue:43493", - "accept": false, - "reason": "Multi-label example-script output bug is unrelated to a SigLIP2 implementation discrepancy." + "reason": "VitPose model breakage vs Ministral-3 FlashAttention failure; different models and failures." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 API messaging and a missing image-patch helper in GLM46V are unrelated." + "reason": "TF32 behavior settings vs missing image-patch helper attribute; unrelated." }, { - "left": "issue:44075", - "right": "issue:44936", + "left": "issue:43116", + "right": "issue:43493", "accept": false, - "reason": "SGD args not being used and `trainer.evaluate()` failing after training are different trainer/optimizer problems." + "reason": "Example script returns empty multi-label results vs SigLIP2 model discrepancy." }, { - "left": "issue:43064", - "right": "issue:43673", + "left": "issue:44075", + "right": "issue:44936", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption and missing GenerationMixin cache during chunked prefill are separate code paths." + "reason": "Optimizer argument handling bug vs trainer.evaluate failure after training; different subsystems." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "Both are Qwen3.5 MoE-related, but one is save/load key serialization and the other is from_pretrained loading failure; not the same bug." + "reason": "Both concern Qwen3.5, but save_pretrained key layout and from_pretrained loading are different concrete bugs." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Different Qwen3 variants and different failures: Qwen3-VL-Moe issue vs Qwen3.5 MoE from_pretrained error." + "reason": "Related model family, but Qwen3-VL-Moe and Qwen3.5-Moe load failures are not the same underlying bug." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Distributed optimizer-state corruption vs missing GenerationMixin cache in chunked prefill." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Empty multi-label example output and a pipeline translation-support error message are unrelated." + "reason": "Classification example misbehavior vs pipeline translation support/error-message issue." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "Qwen3_omni_moe feature-length helper behavior and an AuxRequest PyTorch version check are unrelated." + "reason": "Qwen3_omni_moe helper behavior vs PyTorch version check; unrelated." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "GenerationConfig invalid and stale num_return_sequences warning are different generation issues." + "reason": "Invalid generation config vs stale num_return_sequences warning; different generation-path defects." }, { "left": "issue:44380", "right": "issue:46077", "accept": false, - "reason": "Both involve attention behavior, but the model, backend, and symptom differ: GPT2 scaling ignored vs Gemma4 building causal masks." + "reason": "GPT2 attention scaling ignored vs Gemma4 bidirectional-attention mask construction; different model-specific behaviors." }, { - "left": "issue:33453", - "right": "issue:39401", + "left": "issue:44117", + "right": "issue:45588", "accept": false, - "reason": "Tokenizer loading regression and Qwen3 offset-mapping bug are related only at a very high level; no same underlying defect is evident." + "reason": "Tokenizer mapping None handling vs flash_attention crash on s_aux=None; unrelated code paths." }, { - "left": "issue:44117", - "right": "issue:45588", + "left": "issue:33453", + "right": "issue:39401", "accept": false, - "reason": "Tokenizer mapping fallback bug and flash_attention crashing on `s_aux=None` are unrelated." + "reason": "Tokenizer loading regression vs Qwen3 offset_mapping bug; both tokenizer-related but not the same failure." }, { "left": "issue:43827", "right": "issue:45446", "accept": false, - "reason": "Docs still referencing removed pipeline APIs and an incorrect PyTorch version check are unrelated." + "reason": "Docs still referencing pipeline() vs a runtime version-check bug; unrelated." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer correctness bugs, but they affect different models and different outputs (offset mapping vs general tokenization results)." + "reason": "Different tokenizer bugs for different model families and symptoms; not one duplicate." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "A VRAM leak in multi-threaded Qwen3ForCausalLM usage does not match an isolated reproduction of another issue." + "reason": "VRAM leak in dataloader threading vs isolated reproduction of a different issue; no concrete match." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "Unexpected `_is_hf_initialized` argument in `init_empty_weights` is a different problem from `PretrainedConfig` type checking." - }, - { - "left": "issue:43937", - "right": "issue:44977", - "accept": false, - "reason": "GenerationConfig validation vs FlashAttention generation failure are distinct failures." - }, - { - "left": "issue:43450", - "right": "issue:43986", - "accept": false, - "reason": "Batched video-processor shape bug and missing-`torchvision` crash are different video-related issues." - }, - { - "left": "issue:44466", - "right": "issue:46002", - "accept": false, - "reason": "Serialization of tied weights and NaNs on checkout/init are unrelated." - }, - { - "left": "issue:45405", - "right": "issue:45446", - "accept": false, - "reason": "PEFT minimum-version packaging issue and AuxRequest PyTorch version gating are unrelated." + "reason": "Unexpected init_empty_weights argument vs PretrainedConfig type-checking regression; distinct loading/type issues." } ] }, @@ -13570,52 +13433,55 @@ "issue:42831", "issue:43012", "issue:43066", - "issue:43097", "issue:43122", "issue:43408", "issue:43421", + "issue:43450", "issue:43577", "issue:43582", "issue:43673", "issue:43742", - "issue:43874", "issue:43937", + "issue:43986", "issue:44075", "issue:44117", "issue:44220", "issue:44261", - "issue:44263", "issue:44410", + "issue:44466", "issue:44483", "issue:44484", - "issue:44485", - "issue:44493", "issue:44496", "issue:44829", "issue:44933", - "issue:45071", + "issue:44977", "issue:45081", "issue:45216", "issue:45278", "issue:45310", "issue:45341", + "issue:45405", + "issue:45446", "issue:45464", "issue:45478", "issue:45542", "issue:45561", "issue:45656", - "issue:46172", + "issue:46002", "issue:46178" ], "soft_pairs": [ + "issue:43450|issue:43986", + "issue:43937|issue:44977", + "issue:44466|issue:46002", + "issue:45405|issue:45446", "issue:44117|issue:45081", "issue:45341|issue:45561", - "issue:42222|issue:46178", "issue:38617|issue:41628", + "issue:42222|issue:46178", "issue:44410|issue:45310", "issue:42617|issue:45341", "issue:44829|issue:45216", - "issue:44484|issue:45310", "issue:39401|issue:43421", "issue:42673|issue:43582", "issue:42490|issue:42673", @@ -13623,244 +13489,220 @@ "issue:42371|issue:44261", "issue:44933|issue:45278", "issue:38617|issue:44220", + "issue:44484|issue:45310", "issue:43937|issue:44496", + "issue:43066|issue:43122", "issue:42222|issue:45542", "issue:44483|issue:45464", - "issue:43066|issue:43122", - "issue:42175|issue:42222", "issue:42371|issue:42831", + "issue:42175|issue:42222", "issue:44075|issue:45656", "issue:43673|issue:45341", "issue:44410|issue:45478", - "issue:43012|issue:43577", - "issue:44484|issue:45478", - "issue:43097|issue:44493", - "issue:41628|issue:45071", - "issue:43874|issue:44261", - "issue:44263|issue:44485", - "issue:45071|issue:45542", - "issue:44263|issue:46172" + "issue:43012|issue:43577" ], - "prepared_review_unit_hash": "9e0586c080244a3cbeb989a7c66ff24c394c541216fa020d637dedab42da91d0", + "prepared_review_unit_hash": "e6d2d338ad7c6589274def7a8ad2d0bb476f8addc246a7ca4c57ac5b06e1ce31", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 36964, - "estimated_input_tokens": 9241, - "estimated_eval_tokens": 18738 + "soft_pair_count": 29, + "serialized_chars": 36161, + "estimated_input_tokens": 9041, + "estimated_eval_tokens": 18338 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are unrelated bug reports, with a few soft similarities around import regressions, tokenizer behavior, testing utilities, and model-loading issues. None of the candidate edges look like definite duplicates, so I would keep them separate.", - "confidence": 0.66, - "canonical_issue_reason": "Issue 45278 is the broadest and most representative regression report in the set, centered on widespread import failures after upgrading Transformers.", + "summary": "This cluster is heterogeneous: most items are unrelated single-bug reports, and none of the soft pairs look like true duplicates or the same concrete change. I would not collapse any of the candidates together.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue: the cluster spans separate import/export regressions, tokenizer behavior changes, model-loading errors, backend/config problems, and optimizer/runtime bugs.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45278 is the best overall issue to anchor the cluster because it is generic and closest to several import-related reports, though the cluster is too mixed for a tight canonical fit.", + "best_issue_reason": "No issue is a strong global representative because the soft pairs do not share one underlying code-path problem.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Different bugs: batched video processor shape handling vs missing torchvision causing AutoProcessor load failure." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "GenerationConfig validation error and Qwen3.5 flash-attention generation failure are separate problems." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "Serialization/tied-weight inconsistency is unrelated to NaNs in classification heads after checkout/init." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A release/version pin issue for PEFT is not the same as a PyTorch version check bug in flex_attention." + }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Both are tokenizer/import-adjacent, but one is a mapping lookup bug and the other is a Mistral regex patch crash; different code paths." + "reason": "Generic tokenizer mapping lookup failure and Mistral regex patch crash affect different code paths." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Both mention testing utils, but one is a vague bug and the other is a specific xdist race on captured_info.txt; not the same failure." + "reason": "A minor testing_utils bug is not the same as an xdist race on captured_info.txt." }, { - "left": "issue:42222", - "right": "issue:46178", + "left": "issue:38617", + "right": "issue:41628", "accept": false, - "reason": "Different subsystems and symptoms: vitpose model breakage versus backend-mapping validation for tensorflow_text." + "reason": "These are different missing-import issues for different symbols and modules." }, { - "left": "issue:38617", - "right": "issue:41628", + "left": "issue:42222", + "right": "issue:46178", "accept": false, - "reason": "Both are import errors, but they concern different missing symbols from different modules." + "reason": "Broken VitPose models and tensorflow_text backend mapping errors are unrelated." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Both involve Qwen variants, but one is missing projections in a layer and the other is a from_pretrained MoE loading error." + "reason": "Missing Qwen3Next projections and Qwen3.5 MoE from_pretrained errors are different model-specific failures." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "A script execution problem and a testing_utils bug are not the same underlying issue." + "reason": "A script execution problem is unrelated to a testing_utils bug." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "Flash-attention training degeneration and save_pretrained checkpoint regression are unrelated code paths." - }, - { - "left": "issue:44484", - "right": "issue:45310", - "accept": false, - "reason": "One is about max_shard_size defaults, the other about Qwen3.5 MoE loading; no shared concrete bug." + "reason": "flash_attention_3 training degeneration and Qwen3.5 checkpoint serialization are different issues." }, { "left": "issue:39401", "right": "issue:43421", "accept": false, - "reason": "Both are tokenizer-related, but one is offset_mapping correctness and the other is runtime post-processor updates." + "reason": "Wrong offset_mapping in Qwen3 tokenizer is not the same as post-processor not updating after runtime special-token changes." }, { "left": "issue:42673", "right": "issue:43582", "accept": false, - "reason": "VRAM leakage in a model used across threads is unrelated to an AppleSilicon TypeError in allocator warmup." + "reason": "VRAM leak in multi-threaded dataloaders is unrelated to an Apple Silicon TypeError in caching_allocator_warmup." }, { "left": "issue:42490", "right": "issue:42673", "accept": false, - "reason": "Local path import warnings and multi-thread VRAM leaks are different issues." + "reason": "Local-path import warning and VRAM leak are different symptoms and code paths." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "A model-type warning and a key error while loading a different model are not the same bug." + "reason": "sam3_video/sam3_tracker config mismatch is unrelated to the MobileLLM key error." }, { "left": "issue:42371", "right": "issue:44261", "accept": false, - "reason": "TF32 settings guidance and MLA layernorm epsilon precision are unrelated." + "reason": "TF32 API guidance/config warning and missing rms_norm_eps precision error are distinct problems." }, { "left": "issue:44933", "right": "issue:45278", "accept": false, - "reason": "A missing import from image_utils is too specific to equate with broad import errors after a version upgrade." + "reason": "A nonexistent image_utils import and broad import errors after upgrade are related in theme but not the same concrete bug." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "An ImportError and an fbank feature extraction issue are different failures." + "reason": "ImportError for layer_type_validation is unrelated to the _torch_extract_fbank_features() issue." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "save_pretrained shard-size behavior and Qwen3.5 MoE loading error are different code paths." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "GenerationConfig validation for GLM-5 is unrelated to unrecognized model_type loading." + "reason": "GenerationConfig invalidity is not the same as an unrecognized model missing model_type." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Wrong decoder type in v5 and changed tokenization across versions are related to tokenizers but not the same underlying bug." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "Vitpose model breakage and missing TensorFlow backend handling are different problems." + "reason": "VitPose model breakage is unrelated to a backend error from installing tensorboard without TensorFlow." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Both mention chat/completions and Qwen3.5, but one is request rejection and the other is a streaming inference failure; not enough to treat as one bug." + "reason": "chat/completions request rejection and Qwen3.5 streaming failure are different failures in the API stack." }, { - "left": "issue:43066", - "right": "issue:43122", + "left": "issue:42371", + "right": "issue:42831", "accept": false, - "reason": "Both are tokenizer regressions, but one is decoder type and the other is changed tokenization output; different concrete defects." + "reason": "TF32 behavior warning is unrelated to the FineGrainedFP8 accuracy issue." }, { "left": "issue:42175", "right": "issue:42222", "accept": false, - "reason": "TensorFlow backend packaging and vitpose model breakage do not share a code-path." - }, - { - "left": "issue:42371", - "right": "issue:42831", - "accept": false, - "reason": "TF32 API guidance and FineGrainedFP8 accuracy are distinct precision-related issues." + "reason": "TensorFlow backend missing from a torch install and broken VitPose models are unrelated." }, { "left": "issue:44075", "right": "issue:45656", "accept": false, - "reason": "Both involve optimizers, but one says SGD args are ignored while the other says optimizer.step is called twice under deepspeed; different bugs." + "reason": "SGD args being ignored and optimizer.step being called twice are different optimizer bugs." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Generation cache behavior and a testing_utils bug are unrelated." + "reason": "Generation cache missing in chunked_prefill is unrelated to a small testing_utils bug." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "Same model family, but one is missing layer projections and the other is a from_pretrained MoE load error." + "reason": "Both mention Qwen3.5/Qwen3Next, but the missing projections bug and the from_pretrained error are not the same concrete failure." }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "A PyTorch warning during bfloat16 compile and incorrect dtype propagation in Blip2 are different symptoms." - }, - { - "left": "issue:44484", - "right": "issue:45478", - "accept": false, - "reason": "Save-pretrained shard sizing and Qwen3.5 MoE loading are unrelated." - }, - { - "left": "issue:43097", - "right": "issue:44493", - "accept": false, - "reason": "Removed config API and unexpected position-id keys are not the same bug." - }, - { - "left": "issue:41628", - "right": "issue:45071", - "accept": false, - "reason": "Importing a missing symbol and PretrainedConfig type checking are unrelated regressions." - }, - { - "left": "issue:43874", - "right": "issue:44261", - "accept": false, - "reason": "Missing image-patch count API and MLA layernorm epsilon precision are different model-component issues." - }, - { - "left": "issue:44263", - "right": "issue:44485", - "accept": false, - "reason": "torch.split return handling in a GLM MoE helper is not the same as a RoPE implementation bug." - }, - { - "left": "issue:45071", - "right": "issue:45542", - "accept": false, - "reason": "Type-checking regression in PretrainedConfig and TensorFlow backend detection are unrelated." - }, - { - "left": "issue:44263", - "right": "issue:46172", - "accept": false, - "reason": "These concern different models and code paths: GLM MoE indexing versus deepseek-v4 dtype/transposition logic." + "reason": "A PyTorch compile warning on bfloat16 models is unrelated to Blip2 dtype fields remaining float32 on load." } ] }, @@ -13885,10 +13727,10 @@ "cluster_id": "cluster-10000-430", "nodes": [ "issue:38617", + "issue:41628", "issue:42757", "issue:43012", "issue:43064", - "issue:43066", "issue:43097", "issue:43335", "issue:43408", @@ -13900,7 +13742,6 @@ "issue:43824", "issue:43873", "issue:43874", - "issue:43906", "issue:43931", "issue:43986", "issue:43992", @@ -13916,9 +13757,9 @@ "issue:44410", "issue:44484", "issue:44485", + "issue:44493", "issue:44704", "issue:44829", - "issue:44938", "issue:44987", "issue:45003", "issue:45071", @@ -13927,35 +13768,43 @@ "issue:45372", "issue:45375", "issue:45440", + "issue:45478", + "issue:45542", "issue:45588", "issue:46050", "issue:46123", - "issue:46172", - "issue:46178" + "issue:46172" ], "soft_pairs": [ + "issue:43097|issue:44493", + "issue:41628|issue:45071", + "issue:43874|issue:44261", + "issue:44263|issue:44485", + "issue:45071|issue:45542", + "issue:44263|issue:46172", "issue:44360|issue:46172", "issue:43012|issue:43408", "issue:43441|issue:44829", + "issue:44484|issue:45478", "issue:43931|issue:44410", - "issue:43931|issue:44038", "issue:43335|issue:45071", + "issue:43931|issue:44038", "issue:43673|issue:46123", "issue:43097|issue:43452", "issue:45003|issue:45341", "issue:43986|issue:45341", "issue:44117|issue:44987", "issue:43408|issue:44368", - "issue:43986|issue:44484", "issue:43874|issue:44263", - "issue:43986|issue:45372", - "issue:43531|issue:45341", + "issue:43986|issue:44484", "issue:45440|issue:46172", + "issue:43531|issue:45341", + "issue:43986|issue:45372", "issue:44261|issue:44485", "issue:38617|issue:42757", "issue:43097|issue:46050", - "issue:43064|issue:44315", "issue:43097|issue:45588", + "issue:43064|issue:44315", "issue:44062|issue:44117", "issue:44410|issue:45375", "issue:38617|issue:44351", @@ -13964,212 +13813,264 @@ "issue:43097|issue:43873", "issue:44704|issue:45103", "issue:43986|issue:44117", - "issue:43824|issue:43986", - "issue:44938|issue:46178", - "issue:43066|issue:43906" + "issue:43824|issue:43986" ], - "prepared_review_unit_hash": "6740c1f0ed4fcccbd9ccadbd3d8d1103b1c5e3bd6d8534683497871ab31b07f6", + "prepared_review_unit_hash": "f824affe9cf75f0f1992ce55c6cc1a61142bf34f0bddbb5cd72cb7283bb99527", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 37597, - "estimated_input_tokens": 9400, - "estimated_eval_tokens": 19056 + "soft_pair_count": 38, + "serialized_chars": 39074, + "estimated_input_tokens": 9769, + "estimated_eval_tokens": 19794 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated Transformers issues with only superficial title similarity. None of the soft pairs look like the same underlying bug/change, so all soft edges are rejected. If an anchor is needed, issue 43097 is the strongest representative because it is a concrete v5 regression with clear discussion and inbound references.", - "confidence": 0.88, - "canonical_issue_reason": "Issue 43097 is the best anchor: it describes a specific, user-facing v5 regression with broader relevance than the more niche reports, and it has stronger discussion/inbound activity than most items here.", + "summary": "No safe duplicates stand out here: the soft pairs mostly share broad subsystem keywords (imports, Qwen, flash attention, warnings) but describe different concrete failures. I would keep issue 43873 as the best representative standalone report in the set.", + "confidence": 0.79, + "canonical_issue_reason": "issue:43873 is the strongest standalone bug report in the batch: open, detailed, and the most actively discussed/inbound-referenced item here.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43097 is the most suitable global representative because it is a clear, high-signal Transformers v5 breakage report rather than a narrow model-specific symptom.", + "best_issue_reason": "issue:43873 has the clearest ongoing impact and enough activity to serve as the most useful representative issue for this collection.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Both mention config/model warnings, but one is about a removed config field and the other about unexpected position-id keys; different bug reports." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "Different failures: missing top-level import vs. PretrainedConfig type-checking regression." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different model families and code paths: image-patch helper missing method vs. MLA layernorm epsilon precision issue." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "Both are GLM-related, but one is a DSA indexer split bug and the other is a RoPE implementation discussion; not the same defect." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "PretrainedConfig type checking and TensorBoard/TF backend detection are unrelated bugs." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "Completely different model/code paths: GlmMoeDsaIndexer split behavior vs. DeepSeek-V4 tensor transpose logic." + }, { "left": "issue:44360", "right": "issue:46172", "accept": false, - "reason": "Both mention low-level math/indexing behavior, but they are different models and different failures; not the same bug." + "reason": "A missing ReLU in one GLM indexer path is unrelated to the DeepSeek-V4 comb transpose bug." }, { "left": "issue:43012", "right": "issue:43408", "accept": false, - "reason": "Both are warning-related, but one is a PyTorch precision warning and the other is a model-type mismatch warning; unrelated causes." + "reason": "Both are warnings, but they concern different models and different warning conditions." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Both involve FlashAttention 3, but one is a model loading failure and the other is degraded training behavior; different code paths." + "reason": "Both involve flash attention, but one is a Ministral-3 load failure and the other is degenerate training in sequence classification; not mergeable as one fix." }, { - "left": "issue:43931", - "right": "issue:44410", + "left": "issue:44484", + "right": "issue:45478", "accept": false, - "reason": "Both are Qwen-family model issues, but one is a weight-shape mismatch and the other is missing projections in a different model family." + "reason": "Save shard-size behavior is unrelated to the Qwen3.5-MoE from_pretrained loading error." }, { "left": "issue:43931", - "right": "issue:44038", + "right": "issue:44410", "accept": false, - "reason": "Different model families and different failure modes; too broad to be the same bug." + "reason": "Both are Qwen model issues, but one is a Qwen3-VL weight-shape mismatch and the other is missing projections in qwen3next; different bugs." }, { "left": "issue:43335", "right": "issue:45071", "accept": false, - "reason": "One is a SwitchTransformers config edge case, the other is a generic type-checking regression in PretrainedConfig; not the same issue." + "reason": "A SwitchTransformers config construction bug is unrelated to a PretrainedConfig typing regression." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both mention Qwen3-VL variants, but one is a specific shape mismatch and the other is a broader v5/Qwen3-VL-Moe compatibility report." }, { "left": "issue:43673", "right": "issue:46123", "accept": false, - "reason": "Both are generation/pipeline related, but the failing behavior and affected subsystem are different." + "reason": "Generation cache missing in chunked prefill and mask pipeline final-batch dropping are different subsystems and symptoms." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "One is an API removal/regression, the other is tokenizer/model loading with gguf_file; different underlying problems." + "reason": "Removed config field warnings and GGUF tokenizer/model loading failures are unrelated." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "Different modules and different bug classes; no shared code-path evidence." + "reason": "modeling_utils sys.modules access and a testing_utils bug do not describe the same failure." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "A video-model dependency crash is unrelated to a testing_utils bug." + "reason": "AutoProcessor/torchvision crash and a testing_utils issue are unrelated." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "Both involve loading/tokenizer paths, but the concrete failure and affected assumption differ." + "reason": "A tokenizer mapping None-handling bug is not the same as a model-specific load failure for physical-intelligence/fast." }, { "left": "issue:43408", "right": "issue:44368", "accept": false, - "reason": "Both are warnings about model/config mismatches, but they concern different model families and distinct warning logic." - }, - { - "left": "issue:43986", - "right": "issue:44484", - "accept": false, - "reason": "Video processing dependency handling and max_shard_size behavior are unrelated." + "reason": "Different warnings in different contexts: SAM3 model-type mismatch vs. tie_word_embeddings guidance for Qwen fine-tuning." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Both are GLM-related, but one is an AttributeError in an image processor and the other is a tensor split return-value issue." + "reason": "Different GLM subcomponents and different failures; no shared underlying code-path bug is evident." }, { "left": "issue:43986", - "right": "issue:45372", + "right": "issue:44484", + "accept": false, + "reason": "Missing torchvision error in AutoProcessor is unrelated to shard-size behavior in save_pretrained." + }, + { + "left": "issue:45440", + "right": "issue:46172", "accept": false, - "reason": "Different processor/dependency failure versus mistral_common import breakage; no same bug." + "reason": "Both are DeepSeek-related, but one is a model implementation divergence and the other is a specific tensor operation bug." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to a testing utility bug." + "reason": "Qwen3-MoE sliding_window behavior and a testing_utils bug are unrelated." }, { - "left": "issue:45440", - "right": "issue:46172", + "left": "issue:43986", + "right": "issue:45372", "accept": false, - "reason": "Both are DeepSeek-related, but they describe different model implementations and different failing operations." + "reason": "Video-model loading without torchvision and Gemma 4 processor loading without ReasoningEffort are different import/runtime failures." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "Both are GLM discussions, but one is about missing rms_norm_eps and the other about RoPE implementation; different defects." + "reason": "Layernorm epsilon precision issue and GLM-5 RoPE implementation are not the same bug." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are ImportErrors, but they are for different symbols from different libraries and different breakages." + "reason": "Both are import errors, but they are for different missing symbols in different packages and do not share a single fix." }, { "left": "issue:43097", "right": "issue:46050", "accept": false, - "reason": "A removed embedding-tie flag is unrelated to quantized save/load corruption." + "reason": "A removed config field warning and a quantization save/load breakage are unrelated." }, { - "left": "issue:43064", - "right": "issue:44315", + "left": "issue:43097", + "right": "issue:45588", "accept": false, - "reason": "Training optimizer-state corruption and Liger application during model_init are different training pipeline issues." + "reason": "Tie-embeddings warning/reporting and flash_attention s_aux=None crash are different issues." }, { - "left": "issue:43097", - "right": "issue:45588", + "left": "issue:43064", + "right": "issue:44315", "accept": false, - "reason": "API removal and a flash_attention sinkless-model crash are different code paths." + "reason": "Optimizer state corruption under FSDP2/PEFT is unrelated to Liger Kernel application with model_init." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "Token-related failures, but one is AddedToken argument duplication and the other is TOKENIZER_MAPPING_NAMES returning None." + "reason": "AddedToken constructor argument collision and TOKENIZER_MAPPING_NAMES None-handling are different tokenizer-loading bugs." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Same broad Qwen area, but one is missing projections in qwen3next and the other is a config field omission in Qwen3_5MoeVision; not the same bug." + "reason": "Different Qwen configuration problems: missing attention projections vs. a silently dropped vision config field." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Both are import-related, but they reference different missing symbols and different modules." + "reason": "Both are import errors, but for different APIs/symbols; no evidence they stem from the same regression." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Both mention Qwen3, but one is a model-specific bug and the other is a different architecture/configuration issue." + "reason": "Qwen3-VL-Moe loading trouble and qwen3next missing projections are distinct model-specific defects." }, { "left": "issue:43502", "right": "issue:43992", "accept": false, - "reason": "Local-files-only networking behavior and missing embed_tokens.weight are unrelated loading bugs." + "reason": "Local-files-only network leakage and missing embed_tokens weight on UMT5 loading are unrelated." }, { - "left": "issue:44938", - "right": "issue:46178", + "left": "issue:43097", + "right": "issue:43873", "accept": false, - "reason": "Python 3.14 import/load failure and BACKENDS_MAPPING validation for tensorflow_text are different compatibility issues." + "reason": "A removed config field warning is unrelated to quantization/offloading behavior." }, { - "left": "issue:43066", - "right": "issue:43906", + "left": "issue:44704", + "right": "issue:45103", "accept": false, - "reason": "Tokenizer decoder typing and an isolated reproduction of another issue are not the same concrete bug." + "reason": "AutoProcessor kwargs forwarding and _process_kwargs_parameters crashing under future annotations are different issues." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Video processor loading without torchvision and tokenizer mapping None-handling are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43986", + "accept": false, + "reason": "Qwen2.5-VL import failure and AutoProcessor video-model crash are different symptoms and likely different fixes." } ] }, @@ -14204,7 +14105,6 @@ "issue:43208", "issue:43299", "issue:43352", - "issue:43475", "issue:43502", "issue:43526", "issue:43550", @@ -14216,6 +14116,7 @@ "issue:43824", "issue:43825", "issue:43867", + "issue:43906", "issue:43976", "issue:43992", "issue:44038", @@ -14230,39 +14131,41 @@ "issue:44514", "issue:44655", "issue:44683", - "issue:45276", + "issue:44938", "issue:45290", "issue:45478", "issue:45479", "issue:45698", "issue:45753", - "issue:46077", "issue:46123", "issue:46132", "issue:46172", + "issue:46178", "issue:46209" ], "soft_pairs": [ + "issue:43066|issue:43906", + "issue:44938|issue:46178", "issue:43208|issue:43550", "issue:43992|issue:44484", - "issue:43012|issue:43550", "issue:43825|issue:44655", + "issue:43012|issue:43550", "issue:43012|issue:44368", "issue:42907|issue:44655", "issue:42907|issue:45753", "issue:43352|issue:44380", "issue:44162|issue:46172", "issue:43352|issue:43976", - "issue:44380|issue:45479", "issue:44380|issue:46132", "issue:43792|issue:46123", + "issue:44380|issue:45479", "issue:43064|issue:44380", - "issue:44222|issue:45478", "issue:44483|issue:44514", - "issue:36683|issue:43582", "issue:43208|issue:43526", - "issue:43012|issue:44683", + "issue:36683|issue:43582", + "issue:44222|issue:45478", "issue:42757|issue:43824", + "issue:43012|issue:44683", "issue:43618|issue:43761", "issue:42490|issue:43012", "issue:45698|issue:46172", @@ -14272,224 +14175,222 @@ "issue:44483|issue:45290", "issue:43756|issue:44485", "issue:43066|issue:46209", - "issue:43299|issue:43867", - "issue:36683|issue:43475", - "issue:45276|issue:46077" + "issue:43299|issue:43867" ], - "prepared_review_unit_hash": "70f8f540ce6709fb33e873826782528a146b671e5000b4de370127f20293c190", + "prepared_review_unit_hash": "4169caf30dee068a1611f1c85d7ff5ab6c8e94c561e2ec882881deabb462c5a2", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 36991, - "estimated_input_tokens": 9248, - "estimated_eval_tokens": 18752 + "serialized_chars": 36942, + "estimated_input_tokens": 9236, + "estimated_eval_tokens": 18728 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is heterogeneous: most items are distinct model-, API-, or backend-specific regressions that only share superficial keywords like import/save/load/compile/attention. I don\u2019t see any true duplicate pair among the soft edges.", - "confidence": 0.84, - "canonical_issue_reason": "No single issue is a safe canonical duplicate target: the set spans unrelated bugs across different models, pipelines, imports, saving/loading, and attention backends.", - "canonical_pr_reason": "No pull requests are present in this cluster.", - "best_issue_reason": "issue:45698 is the strongest representative issue here: it is open, user-facing, well-discussed, and describes a concrete save/load regression with broader applicability than the more model-specific reports. It is not, however, a duplicate hub for the whole cluster.", + "summary": "This set is mostly a grab bag of unrelated Transformers issues; despite some superficial overlap (v5 regressions, model families, save/load paths), none of the soft-similarity pairs look like true duplicates or mergeable PR-equivalents.", + "confidence": 0.81, + "canonical_issue_reason": "issue:44483 is the most representative high-impact regression report in the set: a concrete v5 API break with clear user-facing impact and a specific endpoint involved.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44483 is the strongest standalone issue to keep: it is specific, severe, and actionable, with a clear failure mode that is easier to triage than the more model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Different bugs: tokenizer decoder type regression vs an isolated reproduction of another issue; no shared failing code path is evident." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 load failure and a BACKENDS_MAPPING tensorflow_text backend error are unrelated import/config problems." + }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "xLSTM training bugs vs Bamba torch.compile/SDPA failure; unrelated models and code paths." + "reason": "xLSTM small-model training bugs and Bamba torch.compile SDPA failure affect different models and failure modes." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Missing embed_tokens.weight during from_pretrained vs a max_shard_size question; different save/load concerns." + "reason": "Missing embed_tokens.weight during UMT5 loading is unrelated to save_pretrained max_shard_size behavior." }, { - "left": "issue:43012", - "right": "issue:43550", + "left": "issue:43825", + "right": "issue:44655", "accept": false, - "reason": "bfloat16 compile warning vs compile failure on Bamba/SDPA; different symptoms and root causes." + "reason": "Pipeline task-support messaging regression and pipeline serialization failure are different code paths." }, { - "left": "issue:43825", - "right": "issue:44655", + "left": "issue:43012", + "right": "issue:43550", "accept": false, - "reason": "pipeline task-message issue vs saving Pipeline objects; different API surfaces." + "reason": "A bfloat16 precision warning during compile is not the same as a model-specific torch.compile SDPA crash." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "compile precision warning vs tie_word_embeddings warning; not the same bug." + "reason": "PyTorch float32 warning during compile and a tie_word_embeddings config warning are different issues." }, { "left": "issue:42907", "right": "issue:44655", "accept": false, - "reason": "Both involve saving, but one is dequantized model persistence and the other is Pipeline save_pretrained; different objects and code paths." + "reason": "Saving dequantized models and saving Pipeline objects involve different save_pretrained paths." }, { "left": "issue:42907", "right": "issue:45753", "accept": false, - "reason": "Save/dequantization bug vs Qwen3_5 image infinite loop; unrelated." + "reason": "Dequantized save failure and a Qwen3.5 image infinite loop are unrelated." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "FA2 unsupported for Nemotron vs GPT2 attention-scaling ignored; different attention issues." + "reason": "Unsupported Flash Attention 2 warning for Nemotron is unrelated to GPT2 attention scaling being ignored." }, { "left": "issue:44162", "right": "issue:46172", "accept": false, - "reason": "ESM2 broken vs deepseek-v4 transpose bug; unrelated models and failures." + "reason": "ESM2 breakage and a DeepSeek tensor transpose bug are different model-specific defects." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Nemotron FlashAttention support vs Python version compatibility; different problems." - }, - { - "left": "issue:44380", - "right": "issue:45479", - "accept": false, - "reason": "GPT2 attention scaling bug vs zero-loss classification bug; different subsystems." + "reason": "A model capability warning and a Python-version compatibility/import issue are not the same underlying bug." }, { "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "Attention scaling vs AttentionInterface.register behavior change; distinct attention-related regressions." + "reason": "GPT2 backend scaling behavior and AttentionInterface.register side effects are distinct attention-stack changes." }, { "left": "issue:43792", "right": "issue:46123", "accept": false, - "reason": "Whisper run failure vs MaskGenerationPipeline dropping partial-batch results; different pipelines." + "reason": "Whisper load/run failure and MaskGenerationPipeline dropping the final partial batch are unrelated pipeline/model bugs." }, { - "left": "issue:43064", - "right": "issue:44380", + "left": "issue:44380", + "right": "issue:45479", "accept": false, - "reason": "Trainer FSDP/PEFT optimizer-state mismatch vs GPT2 attention scaling; unrelated training vs inference bug." + "reason": "Attention scaling ignored in GPT2 is unrelated to a degenerate zero-loss classification bug." }, { - "left": "issue:44222", - "right": "issue:45478", + "left": "issue:43064", + "right": "issue:44380", "accept": false, - "reason": "FP8 save_pretrained failure vs Qwen3.5 MoE from_pretrained error; one is save-time, the other load-time." + "reason": "Distributed optimizer-state corruption in Trainer/FSDP2 is not the same as GPT2 attention-scaling behavior." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "Chat/completions request rejection vs batched apply_chat_template crash; different request paths." + "reason": "Chat/completions request rejection and chat-template crashes on tool-call messages are different layers of the stack." }, { - "left": "issue:36683", - "right": "issue:43582", + "left": "issue:43208", + "right": "issue:43526", "accept": false, - "reason": "Gemma3Config missing vocab_size vs Apple Silicon allocator TypeError; unrelated." + "reason": "xLSTM training blockers and BeitImageProcessorFast reduce_labels truncation are unrelated." }, { - "left": "issue:43208", - "right": "issue:43526", + "left": "issue:36683", + "right": "issue:43582", "accept": false, - "reason": "xLSTM training blockers vs BeitImageProcessorFast label reduction; different components." + "reason": "Gemma3Config missing vocab_size and Apple Silicon allocator warmup TypeError are completely different failures." }, { - "left": "issue:43012", - "right": "issue:44683", + "left": "issue:44222", + "right": "issue:45478", "accept": false, - "reason": "bfloat16 compile warning vs flex_attention compile failure on torch>=2.9; different compile issues." + "reason": "FP8 save_pretrained problems and Qwen3.5 MoE from_pretrained loading errors are separate issues." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Missing is_offline_mode import vs missing Qwen2_5_VLForConditionalGeneration import; different symbols and causes." + "reason": "Missing is_offline_mode from huggingface_hub and missing Qwen2_5_VLForConditionalGeneration from transformers are unrelated import errors." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "A warning during compilation and a hard flex_attention failure on torch>=2.9 are different backend problems." }, { "left": "issue:43618", "right": "issue:43761", "accept": false, - "reason": "CLIP attentions missing vs CLIPVision hidden_states missing; same family, but distinct regressions and outputs." + "reason": "Both are CLIP regressions, but one is attentions missing and the other is hidden_states missing; they do not look like the same bug." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Trailing-slash local model warning vs bfloat16 compile warning; unrelated." + "reason": "A trailing-slash local model warning is unrelated to the bfloat16 compile warning." }, { "left": "issue:45698", "right": "issue:46172", "accept": false, - "reason": "Wrong custom module loaded after save_pretrained vs deepseek-v4 comb transpose bug; different features." + "reason": "Wrong custom module restoration after save_pretrained and a DeepSeek dtype/transpose crash are different problems." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "ImportError for layer_type_validation vs local_files_only still making requests; unrelated." + "reason": "Importing a removed symbol and local_files_only still making API calls are unrelated import vs network-behavior bugs." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Qwen3-VL-Moe loading breakage vs FP8 inference support request; same model family but different failure modes." + "reason": "Qwen3-VL-Moe loading regression and fp8 inference support are not the same concrete failure." }, { "left": "issue:44483", "right": "issue:45290", "accept": false, - "reason": "chat/completions API rejection vs tool-call template crash; different chat/template paths." + "reason": "Server request acceptance and apply_chat_template crashing on tool-call assistant messages are separate code paths." }, { "left": "issue:43756", "right": "issue:44485", "accept": false, - "reason": "Smollm3 RoPE layer omission vs GLM-5 RoPE implementation discussion; different models and fixes." + "reason": "Smollm3 RoPE layer mismatch and GLM-5 RoPE implementation discussion do not describe the same defect." }, { "left": "issue:43066", "right": "issue:46209", "accept": false, - "reason": "Wrong tokenizer decoder type vs FP8 B200 inference failure; unrelated." + "reason": "Tokenizer decoder type regression and FP8 inference failure on B200 are unrelated." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Qwen3VL-Moe loading breakage vs state_dict-sorting load error; different load-time bugs." - }, - { - "left": "issue:36683", - "right": "issue:43475", - "accept": false, - "reason": "Missing vocab_size on Gemma3Config vs missing fpn_position_embeddings on SAM3 output; different models and attributes." - }, - { - "left": "issue:45276", - "right": "issue:46077", - "accept": false, - "reason": "Gemma4 embedding-resize bug vs Gemma4 bidirectional-attention mask bug; distinct code paths." + "reason": "Qwen3VL-Moe loading breakage and a state_dict sorting load error are different load-time bugs." } ] }, @@ -14514,7 +14415,6 @@ "cluster_id": "cluster-10000-430", "nodes": [ "issue:36010", - "issue:36296", "issue:36683", "issue:41669", "issue:42490", @@ -14525,8 +14425,8 @@ "issue:43064", "issue:43299", "issue:43352", + "issue:43475", "issue:43525", - "issue:43575", "issue:43611", "issue:43618", "issue:43867", @@ -14548,6 +14448,7 @@ "issue:44938", "issue:45071", "issue:45216", + "issue:45276", "issue:45278", "issue:45310", "issue:45357", @@ -14562,6 +14463,8 @@ "issue:46202" ], "soft_pairs": [ + "issue:36683|issue:43475", + "issue:45276|issue:46077", "issue:42994|issue:44655", "issue:43901|issue:44655", "issue:36010|issue:42757", @@ -14570,10 +14473,10 @@ "issue:36683|issue:44077", "issue:43352|issue:43937", "issue:43611|issue:44655", - "issue:43867|issue:45216", "issue:44625|issue:46077", - "issue:43976|issue:44938", + "issue:43867|issue:45216", "issue:46077|issue:46082", + "issue:43976|issue:44938", "issue:43352|issue:45593", "issue:36683|issue:43525", "issue:43618|issue:45216", @@ -14582,8 +14485,8 @@ "issue:43064|issue:44492", "issue:43010|issue:45278", "issue:43992|issue:45310", - "issue:44355|issue:44855", "issue:43867|issue:46202", + "issue:44355|issue:44855", "issue:43867|issue:45375", "issue:41669|issue:44246", "issue:36683|issue:43064", @@ -14591,251 +14494,259 @@ "issue:42890|issue:45561", "issue:43010|issue:45850", "issue:43867|issue:45357", - "issue:43010|issue:43976", "issue:42490|issue:44075", - "issue:43618|issue:44077", + "issue:43010|issue:43976", "issue:43992|issue:45478", - "issue:36296|issue:43575" + "issue:43618|issue:44077" ], - "prepared_review_unit_hash": "4bbfaf0c11ca62ca7e62662170734609e5cb79d2dcba55b34922f169cc56a002", + "prepared_review_unit_hash": "08859d77dc5680a8a0c12269b755af913541fb035eb9cf584d5f1a1c52e3fe61", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 47, "item_count": 47, - "soft_pair_count": 34, - "serialized_chars": 36794, - "estimated_input_tokens": 9199, - "estimated_eval_tokens": 18654 + "soft_pair_count": 35, + "serialized_chars": 37194, + "estimated_input_tokens": 9299, + "estimated_eval_tokens": 18854 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose cluster of mostly unrelated Transformers regressions and loading/import issues. The only commonality is that many are version-regression or model-loading complaints; none of the soft pairs look like true duplicates of the same underlying bug.", - "confidence": 0.93, - "canonical_issue_reason": "issue:45850 is the broadest umbrella report in the set and best captures the general \"new version breaks many models\" theme.", + "summary": "Mostly heterogeneous cluster, but not only one plausible duplicate edge: 41669\u219444246 is a real import-slowdown duplicate candidate, and 45310\u219445478 are an exact-title Qwen3.5 from_pretrained pair that looks duplicate-like, with nearby Qwen3.5 save/load reports forming a loose family. Even so, the rest are largely separate model/config regressions, so the cluster still fits a broad umbrella issue rather than one tight bug.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45278 is still the broadest umbrella: it explicitly reports widespread import failures after a version upgrade, which best matches the cluster\u2019s general regression/import theme. It is broader than the Qwen3.5 mini-family, even though that family contains a duplicate-looking title pair.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 is the most suitable representative issue because it is the most general and can subsume the other scattered regression reports without claiming they are the same bug.", + "best_issue_reason": "issue:45278 is the best global representative because it captures the widest, clearest user-facing regression pattern in the set. The cluster also contains a smaller Qwen3.5 save/load family and an import-slowdown duplicate pair, but they are narrower than 45278\u2019s broad import-breakage report.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Different models and different missing attributes (`vocab_size` vs `fpn_position_embeddings`); not the same bug." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Both are Gemma4-related, but one is token-embedding resizing and the other is attention-mask construction; separate code paths." + }, { "left": "issue:42994", "right": "issue:44655", "accept": false, - "reason": "Both involve saving, but one is quantized model saving failure and the other is saving Pipeline objects; different code paths and failure modes." + "reason": "Both involve saving, but quantized-model saving and Pipeline `save_pretrained` are distinct failure modes." }, { "left": "issue:43901", "right": "issue:44655", "accept": false, - "reason": "Docs mismatch about return_all_scores is unrelated to Pipeline save_pretrained behavior." + "reason": "A docs mismatch and a pipeline serialization bug are not the same underlying issue." }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Different import failures from different packages/modules: transformers.generation vs huggingface_hub." + "reason": "Both are import errors, but they involve different APIs/packages and different breakages." }, { "left": "issue:43299", "right": "issue:45753", "accept": false, - "reason": "One is Qwen3VL MoE model loading regression; the other is an infinite generation loop for a specific image. Not the same bug." + "reason": "Same model family, but one is a loading regression and the other is an image-specific generation loop." }, { "left": "issue:44077", "right": "issue:44380", "accept": false, - "reason": "PatchTSMixer post_init validation and GPT2 attention scaling under SDPA/FlashAttention are unrelated issues." + "reason": "Unrelated model/config problems; no shared concrete code path." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Missing Gemma3Config vocab_size is unrelated to patchtsmixer post_init policy." + "reason": "Different model configs and different symptoms; only superficial similarity around optional config fields." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "Flash Attention 2 support error for Nemotron is unrelated to GLM-5 generation config validation." + "reason": "Flash Attention support gating and invalid `GenerationConfig` are unrelated failures." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "base_model_prefix loading regression and Pipeline save_pretrained are different model-loading vs saving problems." - }, - { - "left": "issue:43867", - "right": "issue:45216", - "accept": false, - "reason": "Sorted state_dict load error is unrelated to Qwen3.5 checkpoint saving regression." + "reason": "Loading with `base_model_prefix` and saving Pipeline objects are distinct serialization behaviors." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "num_labels propagation in Qwen3.5 is unrelated to Gemma4 attention-mask construction." + "reason": "Different models and unrelated bugs (`num_labels` propagation vs attention-mask behavior)." }, { - "left": "issue:43976", - "right": "issue:44938", + "left": "issue:43867", + "right": "issue:45216", "accept": false, - "reason": "Both mention Python compatibility, but one is Python 3.9+ vs 3.10+ behavior and the other is Python 3.14 import failure; not the same defect." + "reason": "Both touch load/save correctness, but state-dict ordering during load and a Qwen3.5 save regression are not clearly the same bug." }, { "left": "issue:46077", "right": "issue:46082", "accept": false, - "reason": "Gemma4 bidirectional attention masking and LlamaConfig head_dim validation are distinct config behaviors." + "reason": "Different Gemma4/Llama config issues with unrelated validation and masking logic." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility complaints, but they target different versions and likely different root causes." }, { "left": "issue:43352", "right": "issue:45593", "accept": false, - "reason": "Flash Attention support gating for Nemotron is unrelated to D-FINE auxiliary loss handling." + "reason": "Separate model features: Flash Attention support vs D-FINE auxiliary-loss handling." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Different missing-config-attribute bugs: Gemma3Config lacks vocab_size, Llama4Config lacks pad_token_id." + "reason": "Missing-config-attribute reports on different models and attributes; too generic to merge." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIPOutput attentions assignment regression is unrelated to Qwen3.5 save_pretrained checkpoint corruption." + "reason": "Different regressions in different models/features; not mergeable into one fix." }, { "left": "issue:36683", "right": "issue:44242", "accept": false, - "reason": "Missing vocab_size on Gemma3Config is unrelated to when load-balancing loss is added for router logits." + "reason": "Unrelated model/config bugs; only both mention configuration behavior." }, { "left": "issue:43352", "right": "issue:43931", "accept": false, - "reason": "Unsupported Flash Attention 2 for Nemotron and Qwen3-VL weight-shape mismatch are different loading/runtime failures." + "reason": "Different model-loading problems with no shared concrete code path." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Trainer/FSDP optimizer-state bug and a typo in cache strategies are unrelated." + "reason": "Training/optimizer-state bug vs a typo in cache strategy names; no overlap beyond being in Transformers." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "Torch no_grad decoration for cache updates and broad import errors after upgrade are not the same code-path problem." + "reason": "A cache-update decoration fix and a broad import-regression report are unrelated." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "UMT5Encoder missing embed_tokens.weight and Qwen3.5 Moe from_pretrained failure are distinct model-loading issues." + "reason": "Both are `from_pretrained` errors, but for different models and different missing/malformed state." }, { - "left": "issue:44355", - "right": "issue:44855", + "left": "issue:43867", + "right": "issue:46202", "accept": false, - "reason": "Compiled Python file errors and a Python 3.13 IndentationError from torch.jit.script parsing are different import/runtime failures." + "reason": "Sorted state_dict load failure and a Gemma pipeline roundtrip issue are different symptoms." }, { - "left": "issue:43867", - "right": "issue:46202", + "left": "issue:44355", + "right": "issue:44855", "accept": false, - "reason": "State_dict sorting load error is unrelated to Gemma pipeline roundtrip failure on the \u2581 character." + "reason": "Both are import-time failures, but compiled Python files and a TorchScript parsing issue are unrelated triggers." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "Sorted state_dict loading and missing deepstack_visual_indexes in Qwen3_5MoeVisionConfig are different bugs." + "reason": "Both are Qwen3.5 checkpoint issues, but state-dict ordering and a missing vision config field are distinct bugs." }, { "left": "issue:41669", "right": "issue:44246", - "accept": false, - "reason": "Removing import * for performance and long import time are related only at a high level, but not the same underlying change or bug." + "accept": true, + "reason": "Both describe slow `import transformers`; one points to wildcard imports as the likely cause and the other reports the same import-time slowdown symptom." }, { "left": "issue:36683", "right": "issue:43064", "accept": false, - "reason": "Gemma3Config attribute omission and FSDP2/PEFT optimizer-state corruption are unrelated." + "reason": "Different failure domains: model config attribute missing vs distributed training optimizer-state corruption." }, { "left": "issue:44938", "right": "issue:45071", "accept": false, - "reason": "Python 3.14 load failure and PretrainedConfig type-checking regression are distinct breakages." + "reason": "Separate Python/runtime compatibility issues with different triggers and symptoms." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Test flakiness from missing set_seed and xdist worker race on captured_info.txt are different concurrency issues." + "reason": "Seed-related flakiness and an xdist file race are different testing bugs." }, { "left": "issue:43010", "right": "issue:45850", "accept": false, - "reason": "Specific cache-update no_grad bug is not the same as the very broad 'breaks too many models' complaint." + "reason": "A specific cache/layer update fix and a vague general complaint about many models breaking are not the same bug." }, { - "left": "issue:43867", - "right": "issue:45357", + "left": "issue:43992", + "right": "issue:45478", "accept": false, - "reason": "Generic load error with sorted state_dict is unrelated to Qwen3.5 incorrect visual encoder keys during save_pretrained." + "reason": "Same broad `from_pretrained` theme, but different models and different reported missing/malformed state." }, { - "left": "issue:43010", - "right": "issue:43976", + "left": "issue:43867", + "right": "issue:45357", "accept": false, - "reason": "Cache update decoration bug and Python version compatibility issue are unrelated." + "reason": "Both are Qwen3.5 checkpoint issues, but one is load failure from ordering and the other is incorrect visual encoder keys on save." }, { "left": "issue:42490", "right": "issue:44075", "accept": false, - "reason": "Import warning for trailing-slash paths is unrelated to optimizer SGD arguments being ignored." - }, - { - "left": "issue:43618", - "right": "issue:44077", - "accept": false, - "reason": "Missing CLIPOutput attentions and patchtsmixer post_init policy are unrelated." + "reason": "Different issues: local-path import warnings vs optimizer SGD arguments being ignored." }, { - "left": "issue:43992", - "right": "issue:45478", + "left": "issue:43010", + "right": "issue:43976", "accept": false, - "reason": "Both are Qwen3.5 loading issues, but one is missing embed token weights and the other is a from_pretrained error; not enough evidence they are the same bug." + "reason": "Cache-update decoration and Python-version support are unrelated." }, { - "left": "issue:36296", - "right": "issue:43575", + "left": "issue:43618", + "right": "issue:44077", "accept": false, - "reason": "Tensor parallel training bug and Qwen2-57B-A14B-Instruct TP OOM are different scaling/load issues." + "reason": "Different regressions in different models/features; not mergeable into one fix." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, + "evaluator_result": { + "accept": false, + "feedback": "The overall read is mostly conservative, but the summary overstates exclusivity: the packet also contains an exact-title duplicate-looking pair, issue:45310 and issue:45478, plus a small Qwen3.5 save/load family that could plausibly cluster together. Because of that, saying the import-time slowdown pair is the only plausible duplicate edge is not fully grounded in the packet." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -14859,7 +14770,6 @@ "issue:33453", "issue:34689", "issue:36296", - "issue:38617", "issue:42548", "issue:42757", "issue:42831", @@ -14869,6 +14779,7 @@ "issue:43232", "issue:43299", "issue:43502", + "issue:43575", "issue:43576", "issue:43618", "issue:43723", @@ -14902,6 +14813,7 @@ "issue:46132" ], "soft_pairs": [ + "issue:36296|issue:43575", "issue:28282|issue:33453", "issue:43618|issue:44857", "issue:44360|issue:45440", @@ -14910,13 +14822,13 @@ "issue:43502|issue:44062", "issue:42890|issue:44112", "issue:42890|issue:45600", - "issue:43576|issue:46132", "issue:43723|issue:44556", + "issue:43576|issue:46132", + "issue:42548|issue:44351", "issue:44387|issue:44556", "issue:42890|issue:45341", - "issue:43010|issue:43065", - "issue:42548|issue:44351", "issue:44222|issue:45125", + "issue:43010|issue:43065", "issue:43502|issue:44485", "issue:44355|issue:44749", "issue:45125|issue:45478", @@ -14927,8 +14839,8 @@ "issue:43502|issue:44222", "issue:43299|issue:44556", "issue:42831|issue:43874", - "issue:43502|issue:44961", "issue:43827|issue:44961", + "issue:43502|issue:44961", "issue:44355|issue:45071", "issue:30333|issue:36296", "issue:44315|issue:44360", @@ -14938,281 +14850,265 @@ "issue:43576|issue:45071", "issue:33453|issue:36296", "issue:43576|issue:43824", - "issue:44222|issue:44230", - "issue:31795|issue:38617" + "issue:44222|issue:44230" ], - "prepared_review_unit_hash": "5c0c165ab6ad377a1d53e8999553459853f81f7c4c4df769877b6fb0fabc8de6", + "prepared_review_unit_hash": "f94c7a8d29b7c02c43e8e7fcdf48bd4ad4a3f40acf1eb7f98485202918003aca", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 38, - "serialized_chars": 38520, - "estimated_input_tokens": 9630, - "estimated_eval_tokens": 19516 + "serialized_chars": 38486, + "estimated_input_tokens": 9622, + "estimated_eval_tokens": 19500 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous cluster of Transformers import/loading/regression reports, with many clearly unrelated edge cases. One Qwen3.5 MoE loading/tensor-parallel pair looks plausibly duplicate; the rest do not.", - "confidence": 0.46, - "canonical_issue_reason": "Issue 44556 is the clearest umbrella for the cluster\u2019s repeated v4\u2192v5 loading-regression theme, with a concrete upgrade/reload failure and broad user impact.", + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues, with a loose theme around v5 loading/regression bugs and model-specific breakages. None of the soft pairs is clearly the same underlying defect, so I would keep them separate.", + "confidence": 0.34, + "canonical_issue_reason": "Issue 43723 is the cleanest representative of the cluster\u2019s common theme: a concrete, user-facing v5 loading regression with a direct API surface and broad relevance.", "canonical_pr_reason": null, - "best_issue_reason": "44556 is the most representative, self-contained issue here and the best anchor for dedup triage across the loading/version-regression subset.", + "best_issue_reason": "43723 is the best anchor issue here because it is specific, actionable, and closest to the repeated \"loading broke in v5\" pattern seen across several items.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Both mention tensor parallelism/loading, but one is a general tensor-parallel training bug and the other is an OOM while loading a specific model. Not the same concrete failure." + }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "PyTorch-missing import dependency vs tokenizer-loading regression; different failure modes and code paths." + "reason": "PyTorch import error versus tokenizer loading regression. Same broad loader area, but different symptoms and code paths." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "CLIP output field regression vs LwDetr float16/CUDA crash; unrelated components." + "reason": "CLIP output attentions bug and LwDETR AMP crash are unrelated model/runtime failures." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer/ReLU discussion vs DeepSeekV3MoE implementation divergence; unrelated bugs." + "reason": "DSA indexer/ReLU discussion is unrelated to DeepSeekV3MoE implementation divergence." }, { "left": "issue:43618", "right": "issue:44977", "accept": false, - "reason": "CLIP attentions assignment vs Qwen3.5 flash-attention generation; different models and symptoms." + "reason": "Different models and failures: missing CLIP attentions versus Qwen3.5 flash-attention generation issues." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "SamHQ test flakiness vs Phi-4 multimodal model-loading failure; not the same underlying issue." + "reason": "Test seeding flakiness is unrelated to a Phi-4 multimodal loading regression." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "local_files_only network leak vs AddedToken keyword conflict; unrelated API bugs." + "reason": "Local-files-only network leakage and AddedToken keyword collision are distinct bugs in different parts of the stack." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Seed-related integration-test flake vs GraniteSpeech CI device-override failure; different tests and causes." + "reason": "One is a missing set_seed in integration tests; the other is a stale device-override CI failure. Not the same bug." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "SamHQ test nondeterminism vs auto_mappings config removal breakage; unrelated." + "reason": "Test seeding issue versus auto_mappings references to removed configs; unrelated failure modes." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both are loading regressions, but one is tokenizer loading and the other is checkpoint reload after upgrade. Too broad to treat as one fix." }, { "left": "issue:43576", "right": "issue:46132", "accept": false, - "reason": "Broken env CLI vs AttentionInterface.register behavior change; distinct surfaces." + "reason": "Broken env command and AttentionInterface.register behavior change are different features and failure paths." }, { - "left": "issue:43723", - "right": "issue:44556", + "left": "issue:42548", + "right": "issue:44351", "accept": false, - "reason": "Tokenizer loading regression vs checkpoint reload after upgrade; both version-related but not the same concrete bug." + "reason": "Both are import errors, but for different symbols and likely different dependency/export issues." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "int4 OOM from reserved memory vs model reload regression; different mechanisms." + "reason": "Int4 CUDA memory regression and checkpoint reload failure are not the same underlying problem." }, { - "left": "issue:43502", - "right": "issue:43065", + "left": "issue:42890", + "right": "issue:45341", "accept": false, - "reason": "Offline-file networking bug vs dummy Conv2d in Sam3PixelDecoder; unrelated." + "reason": "Test utility bug is not the same as the SAM HQ seeding-related integration failure." }, { - "left": "issue:42548", - "right": "issue:44351", + "left": "issue:44222", + "right": "issue:45125", "accept": false, - "reason": "Missing PreTrainedModel export vs missing HybridCache export; similar symptom class, but different symbols/code paths." + "reason": "FP8 save_pretrained bug and missing tensor-parallel plan for Qwen3.5 MoE are separate issues." }, { - "left": "issue:44222", - "right": "issue:45125", + "left": "issue:43010", + "right": "issue:43065", "accept": false, - "reason": "FP8 save_pretrained bug vs missing _tp_plan for tensor parallelism; different features." + "reason": "Cache/layer update no_grad decoration and dummy Conv2d in Sam3PixelDecoder are unrelated." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "local_files_only request leakage vs GLM-5 RoPE implementation; unrelated." + "reason": "Network-request leakage on local_files_only and GLM-5 RoPE implementation are different bugs." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled-file import errors vs performance slowdown after upgrade; not the same bug." + "reason": "Compiled Python file errors and a performance regression after upgrade are not the same issue." }, { "left": "issue:45125", "right": "issue:45478", - "accept": true, - "reason": "Both point to the same Qwen3.5 MoE loading/tensor-parallel regression in v5.4.0; one is the concrete missing-_tp_plan cause, the other the user-facing from_pretrained failure." + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is missing _tp_plan and the other is a broader from_pretrained error in v5.4.0. Not enough evidence of one concrete fix." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLflow reporting omission on failed training vs tokenizer-loading regression; unrelated." + "reason": "MLFlow job reporting and tokenizer loading regression are unrelated." }, { "left": "issue:43010", "right": "issue:43232", "accept": false, - "reason": "Cache update no_grad request vs generation kwargs update after sync_gpus; both generation-related but different code paths." + "reason": "Both are generation-related, but one is cache/layer update decoration and the other is sync_gpus/_update_model_kwargs_for_generation behavior." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Tensor-parallel planning issue vs missing deepstack_visual_indexes config field; different model/config bugs." + "reason": "Both are Qwen3.5 MoE-family items, but one is tensor parallelism and the other is a missing vision config field. Different bugs." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Sentence Transformers loading hang vs documentation confusion about forward args; unrelated." + "reason": "Sentence Transformers loading stall and documentation confusion around forward args are not duplicates." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "local_files_only network bug vs FP8 save_pretrained; different areas." + "reason": "local_files_only network bug and FP8 save_pretrained moe bug are unrelated." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Qwen3VL MoE loading break in v5.0.dev0 vs v4.57\u2192v5.2 reload regression; similar theme, but not the same concrete failure." + "reason": "Different loading regressions on different model families and versions; not a single concrete defect." }, { "left": "issue:42831", "right": "issue:43874", "accept": false, - "reason": "FP8 accuracy issue vs missing image-processor method causing AttributeError; unrelated." + "reason": "FP8 accuracy issue and missing image-processor method causing multimodal token counting failure are distinct." }, { - "left": "issue:43502", + "left": "issue:43827", "right": "issue:44961", "accept": false, - "reason": "API requests despite local_files_only vs unrelated placeholder issue." + "reason": "Documentation references to removed pipeline APIs are unrelated to the placeholder 'racoon' issue." }, { - "left": "issue:43827", + "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "Docs still referencing removed pipeline APIs vs unrelated placeholder issue." + "reason": "No relation between local_files_only networking and the 'racoon' issue." }, { "left": "issue:44355", "right": "issue:45071", "accept": false, - "reason": "Compiled-file errors vs PretrainedConfig type-checking regression; different failures." + "reason": "Compiled-file runtime errors and PretrainedConfig type checking regressions are different problems." }, { "left": "issue:30333", "right": "issue:36296", "accept": false, - "reason": "MLflow/job-status reporting bug vs tensor-parallel training bug; unrelated." + "reason": "MLFlow reporting failure and tensor-parallel training bug are unrelated." }, { "left": "issue:44315", "right": "issue:44360", "accept": false, - "reason": "model_init not applying Liger Kernel vs DSA indexer ReLU discussion; unrelated." + "reason": "Liger Kernel application with model_init and DSA indexer/ReLU discussion do not share a concrete bug." }, { "left": "issue:44355", "right": "issue:44655", "accept": false, - "reason": "Compiled Python file errors vs saving Pipeline objects; separate issues." + "reason": "Running compiled Python files and saving Pipeline objects are separate issues." }, { "left": "issue:33453", "right": "issue:34689", "accept": false, - "reason": "Tokenizer-loading regression vs Llama 3.2 Vision model-loading failure; both load-related but different models/symptoms." + "reason": "Tokenizer loading regression and Llama 3.2 Vision model loading breakage are different model/load failures." }, { "left": "issue:42757", "right": "issue:43576", "accept": false, - "reason": "Missing is_offline_mode import vs broken env CLI; different APIs and code paths." + "reason": "Missing is_offline_mode import from huggingface_hub is not the same as a broken transformers env command." }, { "left": "issue:43576", "right": "issue:45071", "accept": false, - "reason": "Broken env CLI vs PretrainedConfig type checking; unrelated." + "reason": "Broken env command and PretrainedConfig type-checking regression are unrelated." }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "Broken env CLI vs missing Qwen2_5_VLForConditionalGeneration export; unrelated import/CLI issues." + "reason": "Env command failure and Qwen2.5-VL import error are different issues." }, { "left": "issue:44222", "right": "issue:44230", "accept": false, - "reason": "FP8 save_pretrained bug vs FP8 inference support; same broad feature area but different concrete behaviors." - }, - { - "left": "issue:31795", - "right": "issue:38617", - "accept": false, - "reason": "Documentation confusion in model.forward vs import error for layer_type_validation; unrelated." - }, - { - "left": "issue:42890", - "right": "issue:44961", - "accept": false, - "reason": "SamHQ test seed flake vs placeholder issue; unrelated." - }, - { - "left": "issue:42890", - "right": "issue:45341", - "accept": false, - "reason": "SamHQ test flake vs testing_utils bug; different files and symptoms." - }, - { - "left": "issue:44360", - "right": "issue:44556", - "accept": false, - "reason": "DSA indexer/ReLU bug vs checkpoint reload regression; unrelated." + "reason": "FP8 save_pretrained bug and FP8 inference support request are related only by topic, not by the same code-path or fix." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary stays close to the packet: it treats the set as mostly heterogeneous, and the one accepted Qwen3.5 MoE pair is a plausible same-bug duplicate based on the titles. The other verdicts are conservative and don\u2019t overclaim 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 }, @@ -15229,6 +15125,7 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-430", "nodes": [ + "issue:31795", "issue:36296", "issue:38617", "issue:41628", @@ -15249,8 +15146,6 @@ "issue:44230", "issue:44246", "issue:44273", - "issue:44295", - "issue:44297", "issue:44315", "issue:44355", "issue:44393", @@ -15260,6 +15155,7 @@ "issue:44623", "issue:44704", "issue:44908", + "issue:44938", "issue:44961", "issue:45003", "issue:45071", @@ -15280,281 +15176,248 @@ ], "soft_pairs": [ "issue:41628|issue:42548", + "issue:31795|issue:38617", "issue:44961|issue:45468", - "issue:38617|issue:43576", "issue:44355|issue:46178", + "issue:43576|issue:44938", "issue:44315|issue:46209", "issue:46143|issue:46144", "issue:44908|issue:44961", "issue:44485|issue:44961", - "issue:43673|issue:44492", "issue:36296|issue:43901", + "issue:43673|issue:44492", "issue:36296|issue:43526", "issue:45362|issue:45850", "issue:44623|issue:45230", "issue:44273|issue:45542", - "issue:44273|issue:45003", "issue:44038|issue:45125", - "issue:44393|issue:45071", + "issue:44273|issue:45003", "issue:44230|issue:46095", - "issue:44393|issue:45753", + "issue:44393|issue:45071", "issue:43576|issue:44556", "issue:43525|issue:45850", + "issue:44393|issue:45753", "issue:41669|issue:43673", "issue:44556|issue:45092", "issue:43572|issue:45850", - "issue:44492|issue:45850", "issue:42371|issue:44492", + "issue:44492|issue:45850", "issue:45278|issue:45850", "issue:44246|issue:44273", "issue:42890|issue:44908", "issue:44704|issue:44908", "issue:43408|issue:44315", "issue:43976|issue:44393", - "issue:43352|issue:44393", - "issue:44297|issue:45230", - "issue:43576|issue:44908", - "issue:44295|issue:45230", - "issue:44908|issue:45071" + "issue:43352|issue:44393" ], - "prepared_review_unit_hash": "4aca0bc2d96e96401844aeb084b215d29bd11228cba1f466f1331e5d3c417350", + "prepared_review_unit_hash": "4ed796aeebfa94850c87e4b03641e5694089ed24a0b05e05bd4539f3d5129604", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 37, - "serialized_chars": 38108, - "estimated_input_tokens": 9527, - "estimated_eval_tokens": 19310 + "soft_pair_count": 34, + "serialized_chars": 37146, + "estimated_input_tokens": 9287, + "estimated_eval_tokens": 18830 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose bundle of Transformers v5 regression reports, mostly import/export failures and model-specific breakages. The broadest umbrella issue is the general v5 breakage report; the rest are mostly distinct symptoms rather than true duplicates.", + "summary": "This cluster is mostly a mix of unrelated model/feature bugs, with one dominant subtheme around v5 import/export regressions and lazy-loading/import behavior. The clearest representative issue is the broad import-regression report.", "confidence": 0.84, - "canonical_issue_reason": "issue:45850 is the broadest umbrella report: it summarizes the general claim that v5 breaks many models, which matches the cluster\u2019s overall mix of import errors and model regressions better than any single symptom-specific issue.", + "canonical_issue_reason": "issue:45278 best captures the main cluster theme: widespread import failures after upgrading to v5.5.0, which is the most broadly representative and actionable report here.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 is the best representative issue for the cluster because it captures the shared high-level regression theme without being tied to one model or one import path.", + "best_issue_reason": "issue:45278 is the strongest umbrella issue for the import-error/lazy-loading regression subset; the other items are either narrower, model-specific, or unrelated.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "Both are import errors, but they are for different missing symbols (`AutoImageProcessor` vs `PreTrainedModel`) and read like separate export regressions, not one concrete bug." + "reason": "Both are import errors, but they target different symbols and there is no evidence they are the same concrete missing-export bug." }, { - "left": "issue:44961", - "right": "issue:45468", + "left": "issue:31795", + "right": "issue:38617", "accept": false, - "reason": "These are unrelated: a one-word issue title versus a Gemma-4 audio positional encoding bug." + "reason": "Documentation confusion about forward args is unrelated to an ImportError for layer_type_validation." }, { - "left": "issue:38617", - "right": "issue:43576", + "left": "issue:44961", + "right": "issue:45468", "accept": false, - "reason": "One is a missing internal import, the other is a broken CLI command in v5; different code paths and different failures." + "reason": "These are completely unrelated issues; one is a placeholder title and the other is a Gemma audio positional-encoding bug." }, { "left": "issue:44355", "right": "issue:46178", "accept": false, - "reason": "Compiled-Python-file import/runtime errors and a missing backend entry for `tensorflow_text` are different problems." + "reason": "Compiled Python file failures and a tensorflow_text backend mapping error are different failure modes with different code paths." + }, + { + "left": "issue:43576", + "right": "issue:44938", + "accept": false, + "reason": "A broken env command and a Python 3.14 load failure are distinct issues, not the same regression." }, { "left": "issue:44315", "right": "issue:46209", "accept": false, - "reason": "Tensor-parallel model-init behavior and FP8 inference on B200 are separate model/runtime issues." + "reason": "Liger Kernel application in model_init and FP8 inference on B200 are separate model/runtime problems." }, { "left": "issue:46143", "right": "issue:46144", "accept": false, - "reason": "Both involve RoFormer, but one is about kwargs propagation and the other about attention-interface usage; not the same concrete bug." + "reason": "Both touch RoFormer, but one is missing kwargs propagation while the other is about using the attention interface; not the same bug." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs handling and a unrelated placeholder issue are not duplicates." + "reason": "Scheduler kwargs handling is unrelated to the unrelated placeholder issue." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "RoPE implementation discussion and an unrelated issue title are not the same bug." + "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." }, { - "left": "issue:43673", - "right": "issue:44492", + "left": "issue:36296", + "right": "issue:43901", "accept": false, - "reason": "Chunked-prefill cache failure and a cache-strategy typo are different levels of issue and different symptoms." + "reason": "Tensor-parallel training bugs and a documentation mismatch for return_all_scores are not the same underlying problem." }, { - "left": "issue:36296", - "right": "issue:43901", + "left": "issue:43673", + "right": "issue:44492", "accept": false, - "reason": "Tensor-parallel training and pipeline docs mentioning `return_all_scores` are unrelated." + "reason": "A cache-missing runtime bug and a typo in cache strategy docs/code are different issues." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor-parallel training and BeitImageProcessorFast label reduction are different subsystems and failures." + "reason": "Tensor-parallel training and BeitImageProcessorFast label reduction are unrelated." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "A specific Qwen3.5 chat crash is not the same underlying bug as the broad claim that v5 breaks many models." + "reason": "A specific Qwen3.5-35B crash is not the same as a vague report that many models break in v5." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "`processor.save_pretrained` missing files is a concrete serializer bug; the other title is too generic to treat as the same issue." + "reason": "Processor save_pretrained missing files is a concrete bug; the other is a generic, underspecified report." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Generic lazy-loading breakage and a TensorFlow/tensorboard backend error are related at a high level but not the same underlying bug." - }, - { - "left": "issue:44273", - "right": "issue:45003", - "accept": false, - "reason": "Both touch import/lazy-loading mechanics, but one is a general lazy-loading failure and the other is a specific unsafe `sys.modules` access; too different to merge." + "reason": "Lazy loading correctness and an undefined tf backend error from tensorboard-only installs are different bugs." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "A Qwen3-VL-Moe compatibility report and a missing `_tp_plan` for tensor parallelism are different model features/issues." + "reason": "Qwen3-VL-Moe support and missing _tp_plan for tensor parallelism are related to the model family but not the same fixable code-path issue." }, { - "left": "issue:44393", - "right": "issue:45071", + "left": "issue:44273", + "right": "issue:45003", "accept": false, - "reason": "Qwen3-VL bounding-box output behavior and `PretrainedConfig` type checking are unrelated." + "reason": "Lazy loading failures and unsafe sys.modules access are adjacent areas, but the reported bugs are distinct." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "FP8 inference support for Qwen models and a DeepSeek weight-conversion question are unrelated." + "reason": "FP8 inference support for Qwen3-VL and a weight-conversion question for DeepSeekV4 are unrelated." }, { "left": "issue:44393", - "right": "issue:45753", + "right": "issue:45071", "accept": false, - "reason": "Both mention Qwen3-VL/Qwen3_5, but hallucination/bbox output and infinite-loop behavior are different bugs." + "reason": "A Qwen3-VL bounding-box output bug and a PretrainedConfig type-checking break are different problems." }, { "left": "issue:43576", "right": "issue:44556", "accept": false, - "reason": "Broken `transformers env` command and checkpoint reload incompatibility after upgrade are separate v5 regressions." + "reason": "Broken env command behavior and checkpoint reload incompatibility across versions are separate regressions." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "Missing `pad_token_id` on `Llama4Config` is a specific config bug, not the same as the broad multi-model v5 breakage report." - }, - { - "left": "issue:41669", - "right": "issue:43673", - "accept": false, - "reason": "Import-time slowdown from `import *` and a missing cache in chunked prefill are not the same bug." - }, - { - "left": "issue:44556", - "right": "issue:45092", - "accept": false, - "reason": "Both are upgrade incompatibilities, but one is generic checkpoint reload breakage and the other is old remote-code incompatibility with meta initialization." + "reason": "A missing pad_token_id on Llama4Config is a specific config bug, not the same as a broad complaint that many models break." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "A missing `pad_token_idx` in one config class is a specific regression, not the same underlying issue as the broad v5 breakage complaint." + "reason": "Missing pad_token_idx in StableLmConfig is a specific config bug, not the same as the generic v5 complaint." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "A typo in cache strategy names is too specific and unrelated to the general v5 regression umbrella." - }, - { - "left": "issue:42371", - "right": "issue:44492", - "accept": false, - "reason": "TF32 API guidance and a cache-strategy typo are unrelated." + "reason": "A cache-strategy typo is not the same underlying bug as a broad model-breakage report." }, { "left": "issue:45278", "right": "issue:45850", "accept": false, - "reason": "They are thematically close (v5 regressions), but one is specifically about many import errors after an upgrade while the other is a broad complaint about many models; not the same concrete bug." + "reason": "Both are about v5 regressions, but 45278 is a concrete import-error report while 45850 is too vague to be the same specific bug." }, { "left": "issue:44246", "right": "issue:44273", "accept": false, - "reason": "Slow import timing and broken lazy loading are related at a subsystem level but not the same failure." + "reason": "Import slowdown and broken lazy loading are different symptoms and likely different fixes." }, { "left": "issue:42890", "right": "issue:44908", "accept": false, - "reason": "Seed-related test flakiness and scheduler kwargs handling are unrelated." + "reason": "Missing set_seed in tests and an inverse_sqrt scheduler kwargs bug are unrelated." }, { "left": "issue:44704", "right": "issue:44908", "accept": false, - "reason": "`AutoProcessor.from_pretrained` kwarg forwarding and inverse-sqrt scheduler kwargs are different code paths." + "reason": "AutoProcessor cached_file kwargs handling and scheduler kwargs handling are unrelated code paths." }, { "left": "issue:43408", "right": "issue:44315", "accept": false, - "reason": "A SAM3 tracker/model-type mismatch and Liger Kernel not being applied on `model_init` are unrelated." + "reason": "A wrong model-type warning and a Liger Kernel/model_init application bug are not the same issue." }, { "left": "issue:43976", "right": "issue:44393", "accept": false, - "reason": "Python-version compatibility and Qwen3-VL bounding-box output behavior are unrelated." + "reason": "Python-version compatibility and a Qwen3-VL output bug are unrelated." }, { "left": "issue:43352", "right": "issue:44393", "accept": false, - "reason": "Flash Attention support for Nemotron and Qwen3-VL output issues are different model-specific bugs." - }, - { - "left": "issue:44297", - "right": "issue:45230", - "accept": false, - "reason": "Tokenizer save metadata mismatch and a generic bug report are not the same issue." - }, - { - "left": "issue:43576", - "right": "issue:44908", - "accept": false, - "reason": "Broken `transformers env` command and scheduler kwargs handling are different regressions." + "reason": "Flash Attention 2 support for NemotronH and Qwen3-VL bounding-box errors are different model-specific bugs." } ] }, @@ -15582,7 +15445,6 @@ "issue:31795", "issue:33453", "issue:34689", - "issue:38617", "issue:41628", "issue:41669", "issue:42371", @@ -15591,8 +15453,6 @@ "issue:43298", "issue:43502", "issue:43519", - "issue:43525", - "issue:43572", "issue:43576", "issue:43606", "issue:43618", @@ -15603,6 +15463,8 @@ "issue:44261", "issue:44263", "issue:44273", + "issue:44295", + "issue:44297", "issue:44315", "issue:44351", "issue:44360", @@ -15616,6 +15478,7 @@ "issue:44908", "issue:44945", "issue:44998", + "issue:45071", "issue:45230", "issue:45290", "issue:45310", @@ -15628,7 +15491,11 @@ "issue:46172" ], "soft_pairs": [ + "issue:44297|issue:45230", + "issue:44295|issue:45230", + "issue:43576|issue:44908", "issue:41669|issue:44351", + "issue:44908|issue:45071", "issue:41669|issue:45341", "issue:44492|issue:44945", "issue:43976|issue:45850", @@ -15642,8 +15509,8 @@ "issue:44655|issue:44861", "issue:44360|issue:45468", "issue:45230|issue:45310", - "issue:44273|issue:44908", "issue:46018|issue:46172", + "issue:44273|issue:44908", "issue:43519|issue:44485", "issue:43519|issue:44908", "issue:41669|issue:43704", @@ -15662,288 +15529,291 @@ "issue:42913|issue:43066", "issue:44908|issue:44998", "issue:44492|issue:44998", - "issue:44861|issue:45290", "issue:43618|issue:44998", + "issue:44861|issue:45290", "issue:44998|issue:45850", - "issue:44998|issue:45341", - "issue:43525|issue:44861", - "issue:43572|issue:44861", - "issue:31515|issue:38617" + "issue:44998|issue:45341" ], - "prepared_review_unit_hash": "ab379597968e2fe4f2f3aa72ba4ae2e4e732f0081cf8fb9e7daea06f8921b9ad", + "prepared_review_unit_hash": "172b2082a6f6ae6dd14e3314af71ed7601d991bf6d369b4a456dcfe0c1984138", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 41, - "serialized_chars": 39186, - "estimated_input_tokens": 9797, - "estimated_eval_tokens": 19850 + "soft_pair_count": 42, + "serialized_chars": 39485, + "estimated_input_tokens": 9872, + "estimated_eval_tokens": 20000 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is noisy and mostly contains distinct regression reports; I found no strong duplicate pairings. The closest common theme is v5-era loading/import regressions, but the issues still describe different concrete failures.", - "confidence": 0.24, - "canonical_issue_reason": "issue:33453 is the broadest substantive loading-regression report in the set; several nearby items are narrower model/version variants, while many others are unrelated API/docs bugs.", + "summary": "This cluster is mostly heterogeneous: most soft edges connect clearly different bugs, docs issues, or unrelated subsystems. I would keep the earliest concrete Qwen3.5 MoE from_pretrained regression as the representative issue, but I would not merge any of the soft-edge pairs.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45310 is the earliest concrete report of the repeated Qwen3.5 MoE from_pretrained failure and is specific enough to represent that regression thread better than the broader or unrelated items in the cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:33453 is the best representative because it captures the shared loading-regression theme without being overly model-specific.", + "best_issue_reason": "issue:45310 is the best issue to keep as the representative record: it has a concrete model-loading failure, a clear version trigger, and enough detail to be actionable.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "A tokenizer save_pretrained metadata mismatch is not the same as a generic placeholder bug report." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Reading a buffer-registered position_ids tensor is a distinct bug from a generic placeholder report." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "A broken env CLI command and a scheduler-kwargs regression are unrelated code paths." + }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-star performance/refactor issue vs a missing HybridCache export; same general area, different bug." + "reason": "Removing import * for performance is unrelated to a missing HybridCache import." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs handling and PretrainedConfig type checking are different problems." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Unrelated import cleanup/performance report vs a testing_utils bug." + "reason": "A model import cleanup/performance issue 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 different failures." + "reason": "A cache-strategy typo is not the same as incorrect pipeline-parallelism output." }, { "left": "issue:43976", "right": "issue:45850", "accept": false, - "reason": "Python version compatibility breakage vs a broad complaint that v5 breaks many models." + "reason": "Python 3.9 support breakage and a broad 'too many models break' complaint are not the same bug." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Different Qwen model families and different failure modes." + "reason": "Different models and failure modes: Qwen3-VL 2D box output versus Qwen3.5 MoE from_pretrained loading." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "post_init policy change vs an AttributeError in tied-weight key handling." + "reason": "Optional post_init policy for patchtsmixer is unrelated to a tied-weights AttributeError." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Docs still referencing pipeline() is unrelated to the inverse_sqrt scheduler kwargs bug." + "reason": "Docs still referencing pipeline() after removals is unrelated to inverse_sqrt scheduler kwargs." }, { "left": "issue:41628", "right": "issue:41669", "accept": false, - "reason": "Missing top-level AutoImageProcessor export is not the same as the import-* performance regression." + "reason": "AutoImageProcessor import failures and import * cleanup/performance are different import issues." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion about forward arguments vs a tokenizer loading regression." + "reason": "Forward docs confusion is not the same as a tokenizer loading regression." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel application with model_init is unrelated to flash_attention_3 training degeneration." + "reason": "Liger Kernel not applied with model_init and flash_attention_3 training degeneration are different acceleration/training bugs." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken transformers env command is unrelated to lazy loading behavior." + "reason": "Broken env command and lazy loading failure are unrelated." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Pipeline save_pretrained serialization vs tied-weight key AttributeError." + "reason": "Saving Pipeline objects and tied-weight key extraction are different serialization paths." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "DSA indexer ReLU discussion and Gemma-4 audio positional encoding are different code paths." + "reason": "A DSA indexer activation issue is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "Generic bug report vs a specific Qwen3.5 MoE from_pretrained error." + "reason": "A generic placeholder report is not the same as a concrete Qwen3.5 MoE loading regression." }, { - "left": "issue:44273", - "right": "issue:44908", + "left": "issue:46018", + "right": "issue:46172", "accept": false, - "reason": "Lazy loading problem and scheduler kwargs handling are unrelated." + "reason": "DeepSeek shared-expert gating and a comb transpose bug are different model implementation issues." }, { - "left": "issue:46018", - "right": "issue:46172", + "left": "issue:44273", + "right": "issue:44908", "accept": false, - "reason": "DeepSeek-V4 shared-expert gating and a transpose bug in deepseek-v4 are different issues." + "reason": "Lazy loading is unrelated to scheduler argument propagation." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Timestamp calculation in Qwen3VL processor vs GLM-5 RoPE implementation." + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different numeric/model bugs." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Timestamp math in a processor is unrelated to scheduler kwargs handling." + "reason": "Timestamp computation in a processor is unrelated to scheduler kwargs." }, { "left": "issue:41669", "right": "issue:43704", "accept": false, - "reason": "Import cleanup/performance regression vs VRAM leak in multi-threaded dataloader use." + "reason": "Import cleanup/performance is unrelated to a Qwen3 VRAM leak." }, { "left": "issue:31795", "right": "issue:34689", "accept": false, - "reason": "Docs confusion about model.forward args vs a model loading regression for Llama 3.2 Vision." + "reason": "Documentation confusion about forward arguments is unrelated to Llama 3.2 model loading failure." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Slow from_pretrained checkpoint loading and forward-doc confusion are different topics." + "reason": "Slow checkpoint loading and forward-doc confusion are different issues." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak in dataloader threads vs GLM-5 RoPE implementation bug." + "reason": "A VRAM leak and a RoPE implementation discussion are different bugs." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "Different model internals and different concrete failures." + "reason": "GLM-5 RoPE behavior and Gemma-4 audio positional encoding are unrelated model-specific math paths." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching lacking audio-model support vs a weight-conversion question." + "reason": "Audio continuous batching support and a DeepSeek-V4 weight-conversion question are not the same bug." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API guidance is unrelated to a VRAM leak." + "reason": "TF32 API behavior and a Qwen3 VRAM leak are unrelated." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "local_files_only network access bug vs timestamp calculation bug." + "reason": "Unexpected API requests under local_files_only and Qwen3VL timestamp math are unrelated." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Processor timestamp math and processor.save_pretrained missing files are different failures." + "reason": "Processor timestamp calculation and processor.save_pretrained missing files are different failures." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Timestamp calculation bug vs VRAM leak in threaded dataloading." + "reason": "Processor timestamp math and a multi-thread VRAM leak are unrelated." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch in bark-small vs Gemma-4 audio encoding." + "reason": "CPU offload device mismatch and Gemma-4 audio positional encoding are different problems." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return-value bug in GlmMoeDsaIndexer vs Gemma-4 audio positional encoding." + "reason": "A torch.split return-value issue is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "Missing rms_norm_eps in MLA q_a_layernorm vs Gemma-4 audio encoding." + "reason": "Missing rms_norm_eps in MLA q_a_layernorm is unrelated to Gemma-4 audio positional encoding." }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both mention tokenizer v5, but one is general behavior drift and the other is a decoder-type mismatch." + "reason": "Both mention tokenizer v5 changes, but one is a broad behavior difference and the other is a specific decoder-type bug; not the same underlying defect." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Scheduler kwargs handling is unrelated to the generic 'Unemployment' report." + "reason": "A scheduler kwargs regression is unrelated to a nonsense placeholder issue." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "Cache-strategy typo and an unrelated generic report are not duplicates." + "reason": "A cache-strategy typo is unrelated to a nonsense placeholder issue." }, { - "left": "issue:44861", - "right": "issue:45290", + "left": "issue:43618", + "right": "issue:44998", "accept": false, - "reason": "Tied-weight key AttributeError vs chat-template tool-call crash." + "reason": "CLIPOutput attentions regression is unrelated to a nonsense placeholder issue." }, { - "left": "issue:43618", - "right": "issue:44998", + "left": "issue:44861", + "right": "issue:45290", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to the generic report." + "reason": "Tied-weight key extraction and chat template tool-call crashes are different code paths." }, { "left": "issue:44998", "right": "issue:45850", "accept": false, - "reason": "Unrelated generic report vs a broad complaint that v5 breaks too many models." + "reason": "A nonsense title is not the same as a broad model-breakage complaint." }, { "left": "issue:44998", "right": "issue:45341", "accept": false, - "reason": "Unrelated generic report vs a testing_utils bug." - }, - { - "left": "issue:43525", - "right": "issue:44861", - "accept": false, - "reason": "Missing pad_token_id in Llama4Config is unrelated to tied-weight key handling." - }, - { - "left": "issue:43572", - "right": "issue:44861", - "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is unrelated to tied-weight key handling." - }, - { - "left": "issue:31515", - "right": "issue:38617", - "accept": false, - "reason": "Slow checkpoint loading is unrelated to the layer_type_validation import error." + "reason": "A nonsense placeholder issue is unrelated to a testing_utils bug." } ] }, @@ -15972,20 +15842,27 @@ "issue:33453", "issue:36296", "issue:37934", + "issue:38617", "issue:42936", "issue:43240", "issue:43298", + "issue:43525", + "issue:43572", "issue:43668", "issue:43717", "issue:43813", "issue:44031", "issue:44230", "issue:44623", + "issue:44861", "issue:45310", "issue:46018", "issue:46144" ], "soft_pairs": [ + "issue:43525|issue:44861", + "issue:43572|issue:44861", + "issue:31515|issue:38617", "issue:44230|issue:44623", "issue:30333|issue:31515", "issue:31515|issue:36296", @@ -16005,142 +15882,160 @@ "issue:37934|issue:43668", "issue:42936|issue:43668" ], - "prepared_review_unit_hash": "a96235851609dc5cf23fb9001dc68f847f6d8dcef94ed4593eb3ce40b8aa96d3", + "prepared_review_unit_hash": "2aa9b4d7065669e5936d6e4e47353ceac550b4cd5eae2f8325523f8cc36ea52c", "component_budget": { "node_count": 430, "item_count": 430, - "soft_pair_count": 1590, - "serialized_chars": 822920, - "estimated_input_tokens": 205730, - "estimated_eval_tokens": 411716 + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 }, "budget": { - "node_count": 17, - "item_count": 17, - "soft_pair_count": 18, - "serialized_chars": 14810, - "estimated_input_tokens": 3703, - "estimated_eval_tokens": 7662 + "node_count": 21, + "item_count": 21, + "soft_pair_count": 21, + "serialized_chars": 18006, + "estimated_input_tokens": 4502, + "estimated_eval_tokens": 9260 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Heterogeneous cluster of unrelated issue reports; the title overlaps are mostly superficial (loading/pretrained/model-internal wording), with no convincing duplicate or PR match.", - "confidence": 0.86, - "canonical_issue_reason": "No single true canonical duplicate stands out; issue 44623 is the best representative because it is the most concrete, narrow, and actionable bug report in an otherwise mixed set.", + "summary": "This cluster is over-broad and mostly driven by title similarity: it mixes unrelated issues across loading, tokenizers, configs, model internals, training, and inference. None of the suggested soft pairs look like the same underlying bug.", + "confidence": 0.93, + "canonical_issue_reason": "No single canonical issue is a good fit here\u2014the items are too heterogeneous, and the soft edges are only lexical similarity matches rather than true duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "44623 is the strongest standalone issue to anchor the cluster: explicit symptom, self-contained scope, and clear fix direction.", + "best_issue_reason": "If forced to pick one representative, issue 43240 is the broadest concrete runtime-bug report in the set (loss kwargs propagation) and has the most inbound references, but it still does not duplicate the others.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Different bugs: one is missing pad_token_id on Llama4Config, the other is an AttributeError in tied-weight key handling." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "StableLm pad_token_idx regression and _tied_weights_keys list crash are separate config/runtime issues." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Checkpoint loading slowness and a missing import from configuration_utils are unrelated code paths." + }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "Different bugs: fp8 Qwen3-VL/Qwen3.5 inference support vs processor.save_pretrained missing files." + "reason": "FP8 inference support for Qwen models is a different change from processor.save_pretrained missing files." }, { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "Unrelated: MLFlow reporting on failed training vs slow from_pretrained checkpoint loading." + "reason": "MLflow job status reporting on failed training and slow from_pretrained checkpoint loading are unrelated." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Different code paths: checkpoint load performance vs tensor-parallel training bug." + "reason": "One is loading performance, the other is a tensor-parallel training bug; not the same defect." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "Same model family but not the same issue: fp8 inference support vs Qwen3.5 MoE from_pretrained error." + "reason": "FP8 support request and Qwen3.5 MoE from_pretrained failure are different issues, not one duplicate bug." }, { "left": "issue:31515", "right": "issue:33453", "accept": false, - "reason": "Both involve loading, but one is slow checkpoint loading and the other is a tokenizer loading regression." + "reason": "Both involve loading, but one is slow checkpoint loading while the other is a tokenizer-loading regression." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Different features/models: continuous batching audio-model support vs fp8 Qwen3/VL support." + "reason": "Continuous batching for audio models and Qwen3-VL/Qwen3.5 FP8 support are different feature gaps." }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Unrelated model internals: Mamba init_weights usage vs RoFormer attention interface." + "reason": "Mamba/Mamba-2 init_weights behavior and RoFormer attention-interface usage are unrelated architecture-specific issues." }, { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Different architectures and symptoms: Mask2former ignore_value handling vs DeepSeek-V4 shared expert gating." + "reason": "Mask2Former ignore_value handling and DeepSeek-V4 shared-expert gating are different model bugs." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "Type-hint issue in ModernBERTConfig vs tokenizer regex warning regression." + "reason": "ModernBERT type-hint correctness and tokenizer regex warning behavior are unrelated." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Different bugs: unused ignore_value in Mask2former vs kwargs not forwarded to loss calculation." + "reason": "Both touch training/loss plumbing, but ignore_value not used and kwargs not passed to loss are different concrete bugs." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "No shared underlying bug: loss kwargs propagation vs RoFormer attention implementation." + "reason": "Loss-kwargs forwarding and RoFormer attention-interface usage do not describe the same code-path problem." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Unrelated: Mask2former ignore_value handling vs tokenizer warning behavior." + "reason": "Mask2Former ignore_value usage and tokenizer regex warnings are unrelated subsystems." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "Different models and failures; not the same code-path problem." + "reason": "Mask2Former loss handling and RoFormer attention implementation are not duplicates." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "Ignore_value handling and a typo in quantization_operation are unrelated." + "reason": "A model loss-parameter bug and a quantization typo are clearly different issues." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Different scope: loss kwargs forwarding vs ModernBERTConfig type hint." + "reason": "Loss-function kwargs forwarding and an incorrect type hint are not the same bug." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "Type-hint correction vs typo fix; no shared bug." + "reason": "Incorrect type annotation and a typo in a quantization attribute path are unrelated." }, { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Different components and defects: Llama4TextL2Norm behavior vs ModernBERTConfig typing." + "reason": "A question about RMS norm behavior and a ModernBERT type-hint issue are unrelated." }, { "left": "issue:42936", "right": "issue:43668", "accept": false, - "reason": "Different bugs in different models; no duplicate evidence." + "reason": "Mask2Former ignore_value handling and ModernBERT norm_eps typing are different model/config issues." } ] }, @@ -16173,52 +16068,52 @@ "issue:10519|issue:11005", "issue:11005|issue:11307" ], - "prepared_review_unit_hash": "03b8c8feca75cf1bfce6be8500f71a6620527be33dc3ec86d317ce124a11eb41", + "prepared_review_unit_hash": "3db553639ca2a85c89f00059655b735feb34b5b8dc99aa6e23e2f443dfb6000a", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2877, - "estimated_input_tokens": 720, - "estimated_eval_tokens": 1696 + "serialized_chars": 2876, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 2877, - "estimated_input_tokens": 720, - "estimated_eval_tokens": 1696 + "serialized_chars": 2876, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three unrelated feature requests in different areas: token truncation behavior, a learning-rate scheduler-like API, and Wav2Vec2 word boundary offsets. No pair looks like the same underlying bug or change.", - "confidence": 0.97, - "canonical_issue_reason": "No single canonical issue: the three issues address distinct features. If a representative is needed, #10519 is the most concrete and well-scoped request.", + "summary": "The three issues are only superficially similar in wording and appear to be unrelated feature requests: truncation behavior, ReduceLROnPlateau-like scheduling, and Wav2Vec2 word time offsets. None look like duplicates of the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 10519 is the most concrete and self-contained request, with a clearly specified behavior change and scope.", "canonical_pr_reason": null, - "best_issue_reason": "#10519 is the best representative issue because it is the most specific and actionable feature request; the other two are broader or about different subsystems.", + "best_issue_reason": "Issue 10519 is the best representative of the cluster only because it is the most specific and actionable; the other two are unrelated topics.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:10519", "right": "issue:11307", "accept": false, - "reason": "Different features: truncation strategy for tokenization vs Wav2Vec2 word time offsets; no shared code-path or concrete duplicate behavior." + "reason": "Different domains and requested changes: token truncation behavior vs. word boundary timestamps in Wav2Vec2." }, { "left": "issue:10519", "right": "issue:11005", "accept": false, - "reason": "Unrelated requests: token truncation option vs a ReduceLROnPlateau-like scheduler API; only broad similarity as feature requests." + "reason": "One is about truncation direction, the other about learning-rate scheduling functionality; no shared underlying bug or fix." }, { "left": "issue:11005", "right": "issue:11307", "accept": false, - "reason": "Different domains and changes: optimizer/scheduler functionality vs audio model word offset extraction; not the same underlying bug or change." + "reason": "Unrelated feature requests in different subsystems, with no concrete overlap in code path or expected change." } ] }, @@ -16250,12 +16145,12 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "73fd0c090caccfceeb46c241a7b428005801372b2949edb00b55c9984b696a27", + "prepared_review_unit_hash": "c1b07b3e511c524eeed707235846b5c44d606a601726a6325f18536c6bc6201e", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2479, + "serialized_chars": 2477, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -16263,7 +16158,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2479, + "serialized_chars": 2477, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -16272,24 +16167,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These issues are not true duplicates: one is about tokenizer behavior, one about an authentication/error-message problem, and one is a model-loading failure. The only loose commonality is Hugging Face model usage, but the underlying bugs differ.", - "confidence": 0.86, - "canonical_issue_reason": "issue:16618 is the broadest and most central user-facing failure among the three, so it is the best representative issue in this small cluster, even though the others are not actual duplicates.", + "summary": "These issues do not look like duplicates of one another. #16618 is a model-loading failure report, while #15980 is about a bad login/private-model download error message and #13244 is about Tapas tokenization differing from TensorFlow code.", + "confidence": 0.87, + "canonical_issue_reason": "#16618 is the most concrete and broadly representative of the cluster because it is the only issue directly about a failed model load, making it the best anchor for the set even though the other issues are not true duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "issue:16618 is the best global issue candidate because it describes a concrete, common failure path ('Can't load the model for bert-base-uncased') and is more representative of a real reproducible bug than the other two, which are narrower and unrelated.", + "best_issue_reason": "#16618 is the best single issue to represent the cluster because it is the most central user-facing failure case among the three; however, it should not be treated as a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Tokenization mismatch with TensorFlow code is a different bug from a model loading failure for bert-base-uncased." + "reason": "Tapas tokenization mismatch and inability to load bert-base-uncased are different problems affecting different code paths; they do not describe the same underlying bug." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "Bad private-model download/login error handling is a different problem from failing to load a public model." + "reason": "A private-model download/authentication error message is unrelated to a generic 'can't load bert-base-uncased' failure; same subsystem perhaps, but not the same bug." } ] }, @@ -16319,7 +16214,7 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "dad9ae53e9fca75a024e93b4b0efa9c7d99fff9270e34df2aea6846dc10444fe", + "prepared_review_unit_hash": "740361ac90c2811f1e0640057570bf7ae02b8fdafde926bbdd0a292547853382", "component_budget": { "node_count": 2, "item_count": 2, @@ -16341,18 +16236,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one requests Flash Attention 2 support for more architectures, while the other requests GGUF support for more architectures. They share a similar community-contribution framing, but the underlying changes are different.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "These are both community contribution feature requests for expanding model/runtime support, but they target different capabilities: Flash Attention 2 vs GGUF. They do not look like the same underlying bug or change, so they should not be deduplicated.", + "confidence": 0.98, + "canonical_issue_reason": "If a representative issue is needed, #33260 is the better anchor because it is newer and has inbound references, but it is not a duplicate of #26350.", "canonical_pr_reason": null, - "best_issue_reason": "If a representative issue must be chosen, 33260 is slightly stronger because it is newer and has inbound references, but it is still a separate feature request from 26350.", + "best_issue_reason": "#33260 is the better cluster representative only in a loose organizational sense; it is more recent and externally referenced, though it covers a different feature area than #26350.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Both are community-contribution feature requests, but they target different backend capabilities (Flash Attention 2 vs GGUF) and do not look like the same underlying change." + "reason": "Different support requests with different concrete targets: Flash Attention 2 architecture coverage vs GGUF architecture coverage. Same broad theme, but not the same underlying change." } ] }, @@ -16400,124 +16295,124 @@ "issue:43834|issue:44365", "issue:44365|issue:44370" ], - "prepared_review_unit_hash": "fa2756f6d37cdc7dab477b42539b69373e5f2d61cc0a1b19faebf7d5253d37d2", + "prepared_review_unit_hash": "988b9a1040506a9b172fc1ba513f55db238f1c38341126cdde23c3fcbb6718cd", "component_budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 15, - "serialized_chars": 8795, - "estimated_input_tokens": 2199, - "estimated_eval_tokens": 4654 + "serialized_chars": 8789, + "estimated_input_tokens": 2198, + "estimated_eval_tokens": 4652 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 15, - "serialized_chars": 8795, - "estimated_input_tokens": 2199, - "estimated_eval_tokens": 4654 + "serialized_chars": 8789, + "estimated_input_tokens": 2198, + "estimated_eval_tokens": 4652 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are separate docs-translation requests, mostly for different languages, plus a few generic template-like variants. They are related thematically but not the same underlying issue, so none of the soft edges should be merged.", - "confidence": 0.95, + "summary": "These are separate i18n translation-request issues, not duplicates of the same underlying bug/change. The Turkish, Spanish, and Romanian tickets are language-specific, and the generic placeholder-titled issues still read like distinct translation requests rather than one shared fix.", + "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 28936 is the best representative because it is a concrete, fully specified open translation request with substantial discussion; the generic placeholder issues are less descriptive and the other language-specific issues are narrower variants.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Different translation request instances; the titles/body context indicate separate docs-localization tasks rather than one duplicate bug or change." + "reason": "Both are translation requests, but there is no evidence they target the same language or the same concrete work item." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Same broad i18n theme, but not the same underlying request; this looks like a distinct language-specific translation issue." + "reason": "Turkish docs translation is a distinct request from the other issue; same broad category is not enough to merge." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Placeholder-like title similarity is not enough; this is still a separate translation request, not one mergeable duplicate." + "reason": "Different translation ticket; the placeholder title does not show this is the same language or task as the Turkish issue." }, { "left": "issue:27088", "right": "issue:38435", "accept": false, - "reason": "Different target language and separate docs translation task, so they should not be deduplicated." + "reason": "Turkish and Romanian docs translations are separate language-specific issues, not duplicates." }, { "left": "issue:27088", "right": "issue:28936", "accept": false, - "reason": "Both are docs translation requests, but for different languages, so they are not the same underlying issue." + "reason": "Turkish and Spanish docs translations are different requests for different target languages." }, { "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "The generic template-like title is not enough to treat this as the same request; this is a separate translation task." + "reason": "Spanish docs translation is distinct from a generic placeholder issue; no concrete shared change is demonstrated." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Different translation target/request; related topic, but not a duplicate." + "reason": "These are both translation-related, but the available evidence does not show the same target language or same underlying task." }, { "left": "issue:28936", "right": "issue:38435", "accept": false, - "reason": "Spanish vs. Romanian docs translation are distinct requests, so they should remain separate." + "reason": "Spanish and Romanian translation requests are separate issues." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "These appear to be separate localization tickets, not one underlying change." + "reason": "No evidence they are the same translation ticket; only the broad docs-i18n theme overlaps." }, { "left": "issue:38435", "right": "issue:44370", "accept": false, - "reason": "Both are translation-related, but they are not the same concrete issue or language request." + "reason": "Romanian translation is distinct from the placeholder-titled issue; not enough to treat as the same request." }, { "left": "issue:38435", "right": "issue:44365", "accept": false, - "reason": "Different underlying translation task; no evidence they should be merged as duplicates." + "reason": "Romanian docs translation and the placeholder issue are not shown to be the same language-specific task." }, { "left": "issue:38435", "right": "issue:43834", "accept": false, - "reason": "The similarity is only thematic/template-level; the underlying requests are distinct." + "reason": "Separate translation requests; shared subsystem is too broad to merge." }, { "left": "issue:43834", "right": "issue:44370", "accept": false, - "reason": "These look like separate template-based translation tickets, not one duplicate issue." + "reason": "The two placeholder issues may be similar in wording, but there is no concrete evidence they describe the same language or work item." }, { "left": "issue:43834", "right": "issue:44365", "accept": false, - "reason": "High textual similarity, but no proof they are the same language request or same underlying task." + "reason": "Likely template-like translation issues, but the data does not establish the same underlying request." }, { "left": "issue:44365", "right": "issue:44370", "accept": false, - "reason": "Near-identical template wording alone is insufficient; they still appear to be separate translation issues." + "reason": "Both look like generic translation tickets, but not enough detail to conclude they are the same issue." } ] }, @@ -16596,250 +16491,250 @@ "issue:43837|issue:45177", "issue:45322|issue:46133" ], - "prepared_review_unit_hash": "c88b94d3b131948cb71daeb76b776aab97926e3c8e8631c86fb4ad914dc64d7c", + "prepared_review_unit_hash": "64555a2a4072fde57243d5f920b2bf87394e0d0f7b6008d735827e3f9900fe95", "component_budget": { "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21656, - "estimated_input_tokens": 5414, - "estimated_eval_tokens": 11084 + "serialized_chars": 21657, + "estimated_input_tokens": 5415, + "estimated_eval_tokens": 11086 }, "budget": { "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21656, - "estimated_input_tokens": 5414, - "estimated_eval_tokens": 11084 + "serialized_chars": 21657, + "estimated_input_tokens": 5415, + "estimated_eval_tokens": 11086 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are thematically similar model-addition requests, but they cover distinct models/families and modalities. I don\u2019t see any true duplicates or mergeable PR-equivalent changes here, so all soft edges are rejected.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue: the cluster spans unrelated model requests across text, audio, and vision, so there isn\u2019t one underlying bug/change.", + "summary": "This cluster is mostly a set of distinct model-request issues that share a generic \"add model support\" pattern, not the same underlying bug/change. No PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43671 is the most central open request in this set by similarity fan-in, so it is the best single representative of the cluster's broad model-request theme.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a strong global representative for deduplication; they share only a broad \u201cadd model support\u201d theme, not the same fix.", + "best_issue_reason": "issue:43671 is the best global issue representative: open, broadly connected to other request-like issues, and close to the center of the similarity graph, though still not a duplicate of them.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Different model families: EuroBert config vs DEIMv2 architecture." + "reason": "Different requested models (EuroBert vs DEIMv2); only the generic \"add model\" framing overlaps." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "Different requests: EuroBert config vs Qwen3-TTS support." + "reason": "EuroBert and Qwen3-TTS are unrelated model additions." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "Different models entirely: EuroBert vs RF-DETR." + "reason": "EuroBert vs RF-DETR are different architectures and use cases." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are separate model additions, not the same change." + "reason": "RF-DETR and DEIMv2 are separate model requests, not the same change." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "Different modalities and models: RF-DETR vs LFM2.5 Audio 1.5B." + "reason": "Object detection model request vs audio model request; unrelated." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "Object detection model request vs sparse transformer request." + "reason": "Different model families and code paths; only broad request similarity." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR and Qwen3-ASR are unrelated model requests." + "reason": "RF-DETR and Qwen3-ASR are unrelated model additions." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "Different model/task: RF-DETR vs Qwen3-TTS." + "reason": "RF-DETR vs Qwen3-TTS are different model support requests." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR are unrelated model additions." + "reason": "NeoBERT and Qwen3-ASR are unrelated; same generic request theme only." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "Different model families: NeoBERT vs weight-sparse transformer." + "reason": "NeoBERT vs weight-sparse transformer are different concrete model changes." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "Text model request vs audio model request." + "reason": "BERT model request vs audio model request; not the same underlying change." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are distinct model requests." + "reason": "RF-DETR and NeoBERT are different models and code paths." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT vs ModernVBERT: different BERT-like models, not the same underlying request." + "reason": "NeoBERT and ModernVBERT are different model requests despite similar naming." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture and Qwen3-ASR are different model support requests." + "reason": "MiniCPM3 architecture vs Qwen3-ASR are distinct model integrations." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 vs sparse transformer: different concrete models and code paths." + "reason": "MiniCPM3 and the weight-sparse transformer request are unrelated." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are separate model additions." + "reason": "MiniCPM3 architecture and ModernVBERT are separate model additions." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 and Voxtral TTS are unrelated models." + "reason": "MiniCPM3 and Voxtral TTS are different model families and tasks." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are different model requests." + "reason": "DEIMv2 and Qwen3-TTS are unrelated model requests." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT models and Qwen3-ASR target different model families." + "reason": "ModernVBERT and Qwen3-ASR are different model support requests." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are unrelated requests." + "reason": "ModernVBERT and Qwen3-TTS are unrelated." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT models vs generic RoPE BERT-like models: overlap in theme, not the same change." + "reason": "ModernVBERT and generic BERT-like RoPE support are not the same concrete change." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like RoPE support and Voxtral TTS are unrelated." + "reason": "Generic RoPE/BERT-variant request vs Voxtral TTS support; different targets." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Same Qwen3 family, but TTS and ASR are different concrete models/tasks." + "reason": "Same vendor family, but TTS and ASR are distinct models/features and would not be one merged PR." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Different TTS model requests: Qwen3-TTS vs Voxtral TTS." + "reason": "Both are audio-related, but Qwen3-TTS and Voxtral TTS are separate model integrations." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated model additions." + "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated model requests." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "TTS support request vs sparse transformer request." + "reason": "TTS support and a weight-sparse transformer proposal are different code-path changes." }, { "left": "issue:43671", "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio are different audio models." + "reason": "Qwen3-TTS and LFM2.5 Audio are separate model additions." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio are separate audio-model requests." + "reason": "ASR and LFM2.5 Audio are different concrete model support requests." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "ASR model request vs universal perception encoder request." + "reason": "Qwen3-ASR and EUPE are unrelated model architectures/tasks." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR and sparse transformer are not the same code-path problem." + "reason": "ASR support and weight-sparse transformer support are different changes." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "Audio model request vs SAM3-LiteText, different modalities." + "reason": "Audio model request vs SAM3-LiteText; no shared underlying bug/change." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Sparse transformer request and EUPE encoder request are unrelated." + "reason": "Weight-sparse transformer and EUPE are distinct proposals." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Different ASR/TTS model requests: Qwen3-ASR vs Voxtral TTS." + "reason": "Qwen3-ASR and Voxtral TTS are separate audio model integrations." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral TTS are unrelated model additions." + "reason": "NeoBERT and Voxtral TTS are unrelated model requests." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR and DeepSeek-OCR2 are different tasks/models." + "reason": "ASR support and OCR2 support are different model additions." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE and TIPSv2 are distinct vision-language model requests, not the same change." + "reason": "Both are vision-language related, but they target different architectures and objectives, not one duplicate change." } ] }, @@ -16869,7 +16764,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "18a95fe56e70239f20f30e9bf33ba76a15e863f043dd6e8e4d90ee229e9963d2", + "prepared_review_unit_hash": "21c10e9a8e8dccf2a940d514e183b60a4a95424da65636112da8158521c2e6ed", "component_budget": { "node_count": 2, "item_count": 2, @@ -16891,18 +16786,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The issues are about different topics: block decoding vs. DeepSeek v3.2 support. They should not be deduplicated.", + "summary": "These two issues are not duplicates: one is a general request about block decoding, while the other asks for DeepSeek v3.2 support. They concern different features with no shared targets or code path evidence.", "confidence": 0.96, - "canonical_issue_reason": "issue:41084 is a concrete decoding-behavior request and is more specific than the model-support request in issue:43037.", + "canonical_issue_reason": "No canonical issue; the requests are unrelated and do not describe the same bug or feature.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41084 is the more focused, actionable issue for indexing, but it does not overlap with issue:43037.", + "best_issue_reason": "Issue 43037 is the more specific, actionable request, but it is not a duplicate of 41084.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Titles and likely scopes are unrelated: block decoding vs. DeepSeek v3.2 support. No evidence they describe the same underlying bug or change." + "reason": "Different topics: block decoding behavior vs model support request; no evidence they describe the same underlying issue." } ] }, @@ -16932,12 +16827,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "eb947df9d583881aaa43ef2d62b0d6d49d98c124b4cbc6a370c79a8a324d9319", + "prepared_review_unit_hash": "0703ab5d91e0f87bf6b6373538982b9e807704471d679825296fb93497b57fe2", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1583, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -16945,7 +16840,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1583, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -16954,18 +16849,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These two issues are related to Trainer customizability, but they describe different problems: one is about `remove_unused_column` behavior, the other is about FLOPS estimation for custom trainers. They should not be treated as duplicates.", - "confidence": 0.94, - "canonical_issue_reason": null, + "summary": "These are related only at a very high level (both concern Trainer custom behavior), but they target different code paths: one is about `remove_unused_column` and input-column handling, the other is about FLOPs estimation for custom trainers. They should not be treated as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43598 is the cleaner representative because it describes a concrete Trainer customization problem around `remove_unused_column`, with a more specific code-path focus than the FLOPs-estimation request.", "canonical_pr_reason": null, - "best_issue_reason": "If a representative issue is needed, #43598 is the more concrete code-path request because it names a specific Trainer behavior (`remove_unused_column`) and its customizability impact; however, it is not a duplicate of #43600.", + "best_issue_reason": "issue:43598 is the better standalone issue to keep as the cluster representative since it is narrowly scoped and likely easier to act on; issue:43600 addresses a separate instrumentation/metrics concern.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Reject: they concern different Trainer behaviors and different fixes. `remove_unused_column` customizability and FLOPS estimation for custom trainers are related only at a broad subsystem level, not the same underlying bug." + "reason": "Reject: the titles and implied code paths differ. `remove_unused_column` affects data-column pruning/custom trainer behavior, while FLOPs estimation is a separate accounting/metrics feature. They are not the same underlying bug." } ] }, @@ -16995,7 +16890,7 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "7f506d137a54d71d031a1b5cf09a4d387200c564947e974c6955eaeb2562e99b", + "prepared_review_unit_hash": "cd614fae16889f9430e48c54c39c0f50440e604419ff611a08cfe166acaa4890", "component_budget": { "node_count": 2, "item_count": 2, @@ -17017,18 +16912,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are not duplicates: one asks for Qwen3-Omni registration in AutoModel/AutoModelForConditionalGeneration, the other asks for Qwen3.5 AutoModelForSequenceClassification support. They target different model variants and different heads.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "These two issues are both about Qwen model support, but they target different gaps: one requests registration for Qwen3-Omni in generic auto classes, while the other asks for sequence-classification support for Qwen3.5. They are related by ecosystem area, not the same underlying bug or change.", + "confidence": 0.9, + "canonical_issue_reason": "Issue #43684 is the broader and more foundational request: adding Qwen3-Omni registration to AutoModel / AutoModelForConditionalGeneration. It is a direct model-registration gap rather than a single task-specific head.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43684 is the broader registration request and is the most central of the two, but it does not cover the sequence-classification support requested in 44405.", + "best_issue_reason": "Issue #43684 is the better representative of the cluster because it covers the core auto-class registration problem, whereas #44405 is a separate, narrower feature request for AutoModelForSequenceClassification on a different config.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying requests: Qwen3-Omni model registration for AutoModel/AutoModelForConditionalGeneration vs Qwen3.5 sequence-classification support. They do not describe the same bug or change." + "reason": "Different targets and likely different fixes: generic Qwen3-Omni auto-model registration versus sequence-classification support for Qwen3.5. They share a model family theme, but not the same concrete change." } ] }, @@ -17065,76 +16960,79 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "9c821136204fa58aa1da6688bb327dd9b3978287d8a3c65507cd0db972339edb", + "prepared_review_unit_hash": "57e3cef988e4209a66fcb9828c80beb0bf14420181f5a73922e5687d9e861440", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4266, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4266, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "All four issues are too underspecified to confidently deduplicate. Three share the generic title \"huggingface\" but provide no concrete bug details; the \"Fett\" issue is even less clearly related. I would keep them separate for now and avoid merging on title similarity alone.", - "confidence": 0.58, - "canonical_issue_reason": "Issue 43845 is the best available canonical candidate because it is the earliest of the repeated \"huggingface\" reports and is the most representative of the apparent duplicate set, even though the actual problem description is too thin to be certain.", + "summary": "43845, 43846, and 43859 look like a tentative same-title cluster, but the packet only shows title similarity, close timestamps, and soft-similarity scores; that is not enough to confidently label them the same duplicate issue. 43835 looks unrelated. No PRs are present.", + "confidence": 0.54, + "canonical_issue_reason": "issue:43845 is the best available representative for the similar 'huggingface' issues because it is the earliest of that small title-matched set, but the duplicate relationship is only tentative from the available metadata.", "canonical_pr_reason": null, - "best_issue_reason": "43845 is the strongest cluster anchor by chronology and title repetition, but the lack of substantive body text means it is only a provisional canonical issue.", + "best_issue_reason": "issue:43845 is the best overall issue candidate because it is the earliest member of the tentatively related 'huggingface' group and there is no stronger distinguishing evidence for the others.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "The titles do not indicate the same underlying bug, and there is no body detail or target linkage to justify deduplication." + "reason": "Different titles and no body/link evidence tying them to the same underlying report; the metadata does not support a duplicate match." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "No concrete evidence ties \"Fett\" to the \"huggingface\" reports; this looks like weak title similarity rather than the same issue." + "reason": "Different titles and no supporting evidence of a shared bug/report." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Insufficient information to conclude the same bug/change; the titles are unrelated beyond being short, low-signal labels." + "reason": "Different titles and no supporting evidence of a shared bug/report." }, { "left": "issue:43845", "right": "issue:43846", "accept": false, - "reason": "These have the same generic title, but the available content is too sparse to confirm they describe the same underlying problem." + "reason": "Same short title and close timing suggest possible relation, but title similarity alone is not enough to confirm the same underlying issue." }, { "left": "issue:43845", "right": "issue:43859", "accept": false, - "reason": "The repeated title is not enough on its own; without a concrete description, I cannot safely treat them as duplicates." + "reason": "The packet lacks body content or explicit linkage; identical titles and timing make them plausible duplicates, but not confirmable." }, { "left": "issue:43846", "right": "issue:43859", "accept": false, - "reason": "Although the titles match, there is no substantive issue text or target evidence to prove they are the same report." + "reason": "Only title similarity and close timestamps are available, which is insufficient to accept this as the same issue." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, + "evaluator_result": { + "accept": false, + "feedback": "The analysis is a bit too strong for the packet. The only evidence for 43845/43846/43859 is identical titles plus close timestamps and soft-similarity scores; the packet does not provide body content or explicit linkage, and all soft pairs have deterministic_accept=false. It is reasonable to say these three are likely related or may be duplicate reports, but not to state confidently that they are the same duplicate issue. The rejection of 43835 as unrelated is fine. Please soften the cluster/canonical wording and avoid overclaiming duplicate identity from title similarity alone." + }, + "evaluator_used": true, + "retried": true, "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null @@ -17169,88 +17067,88 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "8fd2a768d4ddc87ebd0c889aac403df4de188e74c06cec7ad6a865f97b5704a4", + "prepared_review_unit_hash": "fa76c0604c8faba6242518d32a002cc89de4080776a94ca87a75748c0a47fdfe", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5794, - "estimated_input_tokens": 1449, - "estimated_eval_tokens": 3154 + "serialized_chars": 5788, + "estimated_input_tokens": 1447, + "estimated_eval_tokens": 3150 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5794, - "estimated_input_tokens": 1449, - "estimated_eval_tokens": 3154 + "serialized_chars": 5788, + "estimated_input_tokens": 1447, + "estimated_eval_tokens": 3150 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is almost entirely low-information noise/spam. Two issues share the exact placeholder title, but there is no actionable content or explicit linkage proving they describe the same underlying problem, and the other items look unrelated.", - "confidence": 0.57, - "canonical_issue_reason": "issue:44371 is the earliest exact '' report, so it is the most reasonable representative of the repeated spam-like entries, even though none of the issues contain substantive detail.", + "summary": "This cluster looks like a mix of spam/placeholder issues rather than one real duplicated bug report. The three \"\" issues are similar in title only, while the documentation and \"Google LLC\" items are unrelated low-signal posts.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44371 is the earliest of the repeated \"\" entries, so it is the least-bad anchor if a canonical node is required, but the cluster does not show a real shared issue.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44371 is the best cluster representative by chronology and exact-title match, but it is still not a meaningful canonical issue in the usual sense.", + "best_issue_reason": "issue:44418 is the only item with a non-placeholder title and is the most readable representative, though it still does not look like a true duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Same placeholder title, but there is no body content or explicit evidence that they are the same underlying issue; exact spam labels alone are not enough." + "reason": "Both are just \"\" placeholders with no evidence of the same underlying bug or request beyond identical noise." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Same placeholder title, but no substantive details link them to one concrete bug or request." + "reason": "Identical spam-like titles are not enough to treat them as the same issue; there is no substantive content linking them." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "Different titles and no shared issue-target or filename evidence; the similarity is too generic to treat as a duplicate." + "reason": "One is a spam placeholder and the other is a generic documentation suggestion; they do not describe the same problem." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "The titles are unrelated, and there is no evidence they refer to the same bug or change." + "reason": "A spam placeholder and a \"Google LLC\" promo-style post are unrelated." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "Generic documentation-style wording versus a spam placeholder; no concrete shared problem is visible." + "reason": "Generic documentation suggestion versus spam placeholder; no shared underlying request is visible." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "Generic documentation-style wording versus a spam placeholder; not enough evidence of a shared underlying issue." + "reason": "Generic documentation suggestion versus spam placeholder; not the same issue." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Both are '' placeholders, but without body text or explicit linkage it is unsafe to merge them as the same report." + "reason": "Same placeholder title, but there is no concrete bug/change described to justify duplicate merging." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "The titles do not match semantically, and there is no evidence of the same issue or change." + "reason": "Two different spam-like posts with no evidence of a shared topic or fix." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "The titles do not match semantically, and there is no evidence of the same issue or change." + "reason": "No substantive overlap beyond low-quality content; these are not the same underlying issue." } ] }, @@ -17280,7 +17178,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "bd1c1afa1e18ed2364da25826f4aadb7df4478065ddd61685f9a872c81f107c5", + "prepared_review_unit_hash": "8c3d91dbf076e1d3e4c5c576e91b3e037d1feaa724defdbdfb0f7588f2a93a97", "component_budget": { "node_count": 2, "item_count": 2, @@ -17302,24 +17200,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues have identical titles and similar bodies, so they are plausible duplicates, but the evidence is still limited and the pair is not a deterministic accept. I\u2019d keep #45145 as the canonical issue for the cluster, but reject the soft-edge duplicate claim as too strong on title/content similarity alone.", - "confidence": 0.81, - "canonical_issue_reason": "#45145 is the better canonical issue because it has the same title as #45151 and slightly stronger linkage (1 inbound reference vs. 0), but the duplicate relationship is only tentative from the available evidence.", + "summary": "The two closed issues are likely related and may be duplicates, but the evidence is not strong enough to call them a certain duplicate. They share an identical title and moderate soft similarity, while the packet does not show near-identical bodies or stronger deterministic signals. #45145 remains the better canonical representative because it has the same topic plus the only inbound reference.", + "confidence": 0.84, + "canonical_issue_reason": "#45145 is the best canonical issue because it matches the same title/topic as #45151 and has slightly stronger linkage via one inbound reference.", "canonical_pr_reason": null, - "best_issue_reason": "#45145 is the best global issue representative for this cluster because it is the slightly more connected of the two near-identical reports.", + "best_issue_reason": "#45145 is a reasonable cluster representative for this issue pair because it appears to be the more referenced copy of the same topic.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", "accept": false, - "reason": "Rejecting this soft edge: the only support shown is identical titles and soft similarity, with no explicit links, shared targets, or deterministic acceptance. That is not strong enough to assert the same underlying issue with confidence." + "reason": "The pair has identical titles and a moderate soft-similarity score, but the packet does not provide enough evidence to conclude a clear duplicate with high confidence. The conservative read is that they are probably related, not definitively the same issue." } ] }, "evaluator_result": { "accept": false, - "feedback": "The canonical-issue choice is reasonably grounded in the packet, but the soft-edge verdict is too strong for the evidence shown. The only support is identical titles plus soft similarity (no explicit links or shared targets), and the pair is marked deterministic_accept=false. Please downgrade that duplicate claim to a more conservative, tentative wording or reject the soft edge." + "feedback": "The canonical choice is reasonable, but the duplicate verdict is too strong for the evidence provided. The packet only shows identical titles and a soft_similarity signal (jaccard ~0.63) with deterministic_accept=false; it does not establish \u201cnear-identical\u201d bodies or a \u201cclear duplicate\u201d with high confidence. Please soften the wording and keep the conclusion conservative." }, "evaluator_used": true, "retried": true, @@ -17400,7 +17298,7 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "24384cc7e5d6f6e90bc762e243063cb408ecac0364b72416c152dd4103e380a4", + "prepared_review_unit_hash": "384d10d13b8de50fa77712737a482ee7f4d8bb8632513a6354365435e9a45ae2", "component_budget": { "node_count": 2, "item_count": 2, @@ -17422,18 +17320,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate 4-bit inference bugs: one is a Gemma4/audio-module `torch.finfo()` type error on uint8 weights, the other is a `BitsAndBytesConfig.llm_int8_skip_modules` regression that drops the default `lm_head` exclusion and triggers an assertion. They share the broad inference/quantization area but not the same underlying failure mode.", - "confidence": 0.94, + "summary": "These are two separate 4-bit inference bugs in the same broad area, but they do not look like duplicates: one is a Gemma4 audio-module uint8/torch.finfo TypeError, the other is a BitsAndBytesConfig skip_modules regression that drops the default lm_head exclusion and triggers an AssertionError.", + "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue #45674 is the cleaner, more general regression report for bitsandbytes 4-bit inference behavior; #45672 is a narrower Gemma4/audio-module symptom with a different root cause.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Both involve 4-bit inference, but the failures are different: `torch.finfo()` on uint8 weights in Gemma4 audio modules versus `llm_int8_skip_modules` clearing the default `lm_head` exclusion and causing an assertion. They are not the same underlying bug." + "reason": "Different failure modes and code paths: 45672 is Gemma4/audio-module dtype handling during NF4 inference, while 45674 is a config regression around llm_int8_skip_modules clearing the default lm_head exclusion. Similar 4-bit context, but not the same underlying bug." } ] }, @@ -17463,7 +17361,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "8128b4b1414626d99d91ea5e998f04957af3c534f334b9763b74fc6b699801a8", + "prepared_review_unit_hash": "46b0a9a374a692058f62728ea176bd577ccdf2d8af148f95419c3c6bfe27b37e", "component_budget": { "node_count": 2, "item_count": 2, @@ -17485,18 +17383,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two closed issues appear to be unrelated low-information name/date entries, not duplicate bug reports or feature requests. No canonical representative is warranted for the pair.", + "summary": "Both issues appear to be unrelated spam/personal-name entries with no technical content; they should not be clustered as duplicates.", "confidence": 0.98, - "canonical_issue_reason": null, + "canonical_issue_reason": "Issue 45762 is marginally earlier, but neither issue has substantive technical detail to serve as a meaningful canonical report.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "If one must be retained as a representative, 45762 is slightly better only because it was created first; otherwise neither is suitable as a canonical issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Both are just distinct personal-name/date titles with no shared problem statement, code reference, or evidence of the same underlying issue." + "reason": "The titles are different personal names with different dates and there is no shared bug, feature, or code-path context; this is not the same underlying issue." } ] }, @@ -17526,7 +17424,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "51431bdc8236ffca501e054a34fbc514375e5caf086da7ba6459bd59c4f6e1d9", + "prepared_review_unit_hash": "68e86a5578e7ed3f85cac345652c5d012325157b2154cb54fd8c996f86bdafc6", "component_budget": { "node_count": 2, "item_count": 2, @@ -17548,18 +17446,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are just closed records with different personal-name/date titles and no shared context, so they do not look like duplicates or part of the same bug report.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "Both items are closed issues with only personal-name/date titles and no body or technical context; they do not appear to describe the same bug or change. This looks like unrelated spam/miscellaneous entries rather than a duplicate cluster.", + "confidence": 0.98, + "canonical_issue_reason": "Neither issue has enough technical content to serve as a canonical bug report; they are just name/date titles and appear unrelated.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "No issue in this pair is a good canonical candidate because there is no shared underlying problem to preserve.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "Titles refer to different people and different dates, with no body, links, or shared evidence of the same underlying issue." + "reason": "The titles refer to different personal names and dates, with no description, links, or evidence of a shared bug/change; not a duplicate." } ] }, @@ -17599,91 +17497,91 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "9b7836abefe48efad78a11e0eb4b228d3779e9cdd520c3afbc52c1879ad18e50", + "prepared_review_unit_hash": "a720d5be17ae9edb80637cf8fdb696adba8ff2244c3724003a6318daa5c2971d", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5726, - "estimated_input_tokens": 1432, - "estimated_eval_tokens": 3120 + "serialized_chars": 5722, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5726, - "estimated_input_tokens": 1432, - "estimated_eval_tokens": 3120 + "serialized_chars": 5722, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two issue pairs look like likely duplicates from identical titles and matching body lengths: 45870/45871 and 45872/45873. 45869 is only thematically related to the financial-data cluster and should stay separate. No PRs are present.", - "confidence": 0.88, - "canonical_issue_reason": "45870 is the better canonical representative for its pair because it matches 45871 on title and body length and has slightly more discussion activity, so it is the more established duplicate target.", + "summary": "This cluster is not one single duplicate set: it contains two exact duplicate issue pairs (45870/45871 and 45872/45873) plus one separate, related-but-distinct issue (45869).", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45870 is the best canonical issue for its pair: it is an exact duplicate of 45871, appears slightly earlier, and has the clearest self-contained title. Note that 45872/45873 form a separate duplicate pair, not the same underlying request.", "canonical_pr_reason": null, - "best_issue_reason": "45870 is the strongest overall issue to keep as canonical: it is part of a likely duplicate pair, has the most engagement among the similar issues, and is more representative than the unrelated dataset-promotion issue.", + "best_issue_reason": "45870 is the best issue representative overall because it is an exact duplicate pair leader and the earliest clear duplicate in the set, but it should not absorb the unrelated 45869 or the separate 45872/45873 request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Shared financial-domain wording, but the titles point to different kinds of issues: a dataset promotion versus an example request." + "reason": "Both mention financial data/time series, but 45869 is a dataset bundle offer and 45870 is an example request; different underlying issue." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Only broadly related by topic; the dataset-advertisement issue does not match the example-request issue." + "reason": "Same as 45869 vs 45870: related topic, but a dataset offer is not the same as an example request." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Same title and matching body length make this a likely duplicate pair, though the body text itself is not visible here." + "reason": "Exact same title and very likely same body/content; these are clear duplicates of the same issue." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Both are DataForge-related, but the titles describe different requests: an example issue versus a scraping-service request." + "reason": "Both are DataForge financial/time-series items, but one is an example request and the other includes a scraping service; different requests." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Same broad subsystem, but the titles suggest different concrete requests, so this is not the same underlying issue." + "reason": "Same as 45870 vs 45872; similar theme, but not the same concrete issue." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Related by domain only; the titles indicate different issue types and likely different requests." + "reason": "Despite shared keywords, these describe different asks: example request vs time-series+scraping service." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Broad topical overlap is not enough here; the title indicates a different concrete request than the example issue." + "reason": "Same as 45871 vs 45872; not the same underlying request." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Same title and matching body length make this a likely duplicate pair, based on the available metadata." + "reason": "Exact same title and matching content pattern; clear duplicate issues." } ] }, "evaluator_result": { - "accept": false, - "feedback": "Reject: the conclusions are plausible, but several reasons are too strong for the evidence in the packet. We only have identical titles and matching body lengths, not the actual bodies, so wording like \"identical content,\" \"straight duplicate pair,\" and \"clear duplicate\" overstates what can be proven. A more conservative read would be that 45870/45871 and 45872/45873 are likely duplicates based on title/body-length match, while 45869 is only thematically related. The no-PR statement is grounded." + "accept": true, + "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 are clearly supported as duplicate pairs by identical titles and matching bodies, while 45869 is only loosely related and should stay separate. The soft-edge verdicts are conservative and do not overstate evidence." }, "evaluator_used": true, - "retried": true, + "retried": false, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -17691,7 +17589,7 @@ { "label": "pull_request", "component_index": 1, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17717,8 +17615,8 @@ ], "soft_pairs": [ "pull_request:39895|pull_request:43424", - "pull_request:41855|pull_request:43823", "pull_request:41855|pull_request:43424", + "pull_request:41855|pull_request:43823", "pull_request:42668|pull_request:43823", "pull_request:43251|pull_request:43823", "pull_request:43251|pull_request:44827", @@ -17726,8 +17624,8 @@ "pull_request:43424|pull_request:43823", "pull_request:43823|pull_request:45655", "pull_request:43823|pull_request:43995", - "pull_request:43995|pull_request:44827", "pull_request:43995|pull_request:45213", + "pull_request:43995|pull_request:44827", "pull_request:44827|pull_request:45170", "pull_request:44827|pull_request:46048", "pull_request:44827|pull_request:45655", @@ -17735,12 +17633,12 @@ "pull_request:43823|pull_request:45213", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "df0cfd57683c353d8cf2c094c46f35b563f9dbdc06d7c7099454504192a9ae38", + "prepared_review_unit_hash": "9c82aec2a7de057e1023f03e9842db4134d385deffb6a782d7843aabf89e610c", "component_budget": { "node_count": 13, "item_count": 13, "soft_pair_count": 18, - "serialized_chars": 34771, + "serialized_chars": 34769, "estimated_input_tokens": 8693, "estimated_eval_tokens": 17642 }, @@ -17748,7 +17646,7 @@ "node_count": 13, "item_count": 13, "soft_pair_count": 18, - "serialized_chars": 34771, + "serialized_chars": 34769, "estimated_input_tokens": 8693, "estimated_eval_tokens": 17642 }, @@ -17757,126 +17655,126 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly heterogeneous PRs across unrelated models, docs, tests, and utilities. The only true near-duplicate pair is the two `fixed_cross_entropy` patches, which target the same helper and the same issue; everything else should be kept separate.", - "confidence": 0.95, + "summary": "Only one true duplicate pair stands out: the two `fixed_cross_entropy` PRs (43251 and 43254). The rest are unrelated model additions, refactors, or tests and should stay separate.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43251 is the clearest representative of the only duplicate pair: it explicitly targets issue 43240 and fixes the same `fixed_cross_entropy` / `nn.functional.cross_entropy` path as 43254, with only minor API-shape differences.", + "canonical_pr_reason": "PR 43251 is the clearest representative of the actual fix: it directly addresses the cross-entropy kwargs bug, has an explicit issue link, and makes the narrowest self-contained change in `loss_utils.py`.", "best_issue_reason": null, - "best_pr_reason": "PR 43251 is the best canonical PR candidate because it directly addresses the concrete loss-wrapper bug in the shared code path; the other PRs are unrelated feature additions, model scaffolding, or test-only changes.", + "best_pr_reason": "PR 43251 is the best overall representative because it fixes the concrete bug most explicitly and with the most direct traceability to the reported issue; PR 43254 is effectively the same change but less specific.", "soft_edge_verdicts": [ { "left": "pull_request:39895", "right": "pull_request:43424", "accept": false, - "reason": "Unrelated PRs: VideoPrism model/docs addition vs. an ExecuTorch dynamic-shape test. Different code paths and no shared bug." + "reason": "Unrelated features: VideoPrism model/docs vs an ExecuTorch dynamic-shape test. Same broad area of training/runtime infrastructure is not enough." }, { "left": "pull_request:41855", - "right": "pull_request:43823", + "right": "pull_request:43424", "accept": false, - "reason": "Different changes: Mistral tokenizer method additions vs. adding MobileLLM model support. Same broad area of Transformers, but not the same underlying fix." + "reason": "Mistral tokenizer methods vs ExecuTorch export test; different code paths and different problems." }, { "left": "pull_request:41855", - "right": "pull_request:43424", + "right": "pull_request:43823", "accept": false, - "reason": "Tokenizer API fix vs. ExecuTorch export test. No common concrete bug or mergeable overlap." + "reason": "Tokenizer API cleanup vs adding a new MobileLLM model; unrelated changes." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "Processor robustness changes across auto/processing utilities vs. a new MobileLLM model. Not the same issue." + "reason": "Processor-pretrained robustness work vs new MobileLLM model implementation; not the same bug/change." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Loss-function kwargs fix vs. MobileLLM model addition. Different subsystems and unrelated intent." + "reason": "Cross-entropy kwargs fix vs new MobileLLM model addition; no shared underlying defect." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "Cross-entropy helper fix vs. Mistral4 test/integration work. Same general repo area, but no shared underlying bug." + "reason": "Loss utility kwargs fix vs Mistral4 test/model updates; different concrete issues." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "ExecuTorch export test vs. Mistral4 test/model changes. Both are tests-heavy, but they do not fix the same problem." + "reason": "ExecuTorch dynamic-shape test vs Mistral4 test fixes; both are tests but not the same underlying change." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "ExecuTorch dynamic-shape test and MobileLLM model addition are unrelated changes." + "reason": "Dynamic-shape export test vs MobileLLM model addition; unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "New MobileLLM model support vs. `cls.config` resolution ordering fix. No shared bug or code path." + "reason": "Model implementation work vs a core class attribute-resolution fix; different code paths and intents." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM model addition vs. Falcon output-collection refactor. Different models and different concerns." + "reason": "New model addition vs Falcon output-collection refactor; not the same change." }, { "left": "pull_request:43995", - "right": "pull_request:44827", + "right": "pull_request:45213", "accept": false, - "reason": "Falcon interface refactor vs. Mistral4 test fixes. Not the same concrete change." + "reason": "Falcon refactor vs a model-creation skill PR; no shared bug or code path." }, { "left": "pull_request:43995", - "right": "pull_request:45213", + "right": "pull_request:44827", "accept": false, - "reason": "Falcon refactor vs. a model-creation skill/documentation PR. Clearly unrelated." + "reason": "Falcon output-interface refactor vs Mistral4 tests/model changes; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test/model fixes vs. a global `layrnorm`\u2192`layernorm` renaming patch. Different bugs and different scopes." + "reason": "Mistral4 tests/model fixes vs `layrnorm` renaming cleanup; different concrete changes." }, { "left": "pull_request:44827", "right": "pull_request:46048", "accept": false, - "reason": "Mistral4 fixes vs. TDT loss kernel integration. No shared underlying code-path problem." + "reason": "Mistral4 test fixes vs adding a TDT loss kernel; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45655", "accept": false, - "reason": "Mistral4 work and `cls.config` resolution ordering are unrelated." + "reason": "Mistral4 test fixes vs config-resolution ordering fix; different code paths." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "Mistral4 PR vs. a model-creation skill PR. Not comparable as duplicates." + "reason": "Model/test fixes vs model-creation skill content; not a duplicate change." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "New MobileLLM model support vs. a skill/docs PR. No shared implementation bug." + "reason": "MobileLLM model implementation vs model-creation skill/docs; unrelated." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs patch `src/transformers/loss/loss_utils.py` to extend `fixed_cross_entropy` for the same issue 43240; the diffs are functionally overlapping and could plausibly be merged into one PR." + "reason": "Both PRs modify `src/transformers/loss/loss_utils.py` to fix the same `fixed_cross_entropy` issue by forwarding supported `nn.functional.cross_entropy` kwargs (notably `weight` and `label_smoothing`). They are mergeable into one PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "The evaluation is grounded in the packet. The only accepted pair, PR 43251 vs 43254, is the same `loss_utils.py` bugfix for issue 43240 and looks mergeable into one PR. The other pair rejections are conservative and consistent with the diffs, which are mostly unrelated model additions, test-only changes, or refactors." + "feedback": "The judgments stay grounded in the packet and the lone accepted duplicate pair is well supported by shared file/issue-target evidence. The other pair rejections are conservative; none look mergeable as a single PR for the same bugfix." }, "evaluator_used": true, "retried": false, @@ -17887,7 +17785,7 @@ { "label": "pull_request", "component_index": 2, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17903,7 +17801,7 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "8c40fe73cc05fc266a48638109c1a55489292fcb1ccc0993eb22cab9615532fd", + "prepared_review_unit_hash": "926984a8d5e460a648e89ef415d1c57fed30b3445d0140979f17e0d00e71766f", "component_budget": { "node_count": 2, "item_count": 2, @@ -17925,18 +17823,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related VibeVoice PRs, but they are not duplicates: one is the base VibeVoice implementation and the other adds a separate realtime path plus an acoustic tokenizer package.", - "confidence": 0.86, + "summary": "Two related VibeVoice PRs, but they target different work: initial VibeVoice support vs. a later realtime/acoustic-tokenizer extension. Not duplicates.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #40546 is the broader, non-draft core VibeVoice implementation and is the better anchor for this cluster; PR #42781 looks like a later follow-on for realtime support rather than the same change.", + "canonical_pr_reason": "PR 40546 is the core VibeVoice implementation (model, docs, auto-registry, processing). PR 42781 is a separate realtime follow-up with additional tokenizer/feature-extractor work.", "best_issue_reason": null, - "best_pr_reason": "PR #40546 is the best representative because it implements the main VibeVoice integration end-to-end and has active review, while #42781 is a narrower draft add-on focused on realtime/tokenizer support.", + "best_pr_reason": "PR 40546 is the better cluster anchor because it represents the main VibeVoice integration; 42781 is a distinct extension rather than the same change.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "They share the VibeVoice subsystem and some filenames, but the scopes differ: #40546 implements VibeVoice generally, while #42781 adds realtime-specific functionality and an acoustic tokenizer. That is a follow-on feature, not the same concrete change." + "reason": "Related but not the same underlying change: 40546 adds base VibeVoice support, while 42781 adds VibeVoice Realtime and acoustic-tokenizer changes. Too distinct to merge as one PR." } ] }, @@ -17950,7 +17848,7 @@ { "label": "pull_request", "component_index": 3, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -17966,7 +17864,7 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "1724e52f04823a6d0b9dc71d18e353482f4ba67f36e3bd0a18686180f5386c92", + "prepared_review_unit_hash": "a9c041c2bb68f8d6a62eb7ce3c9a258214f1488d3d437c2cc1a47abef4d421b8", "component_budget": { "node_count": 2, "item_count": 2, @@ -17988,18 +17886,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These pull requests touch entirely different models and code paths: TimesFM covariate forecasting vs. DINOv3 image classification support. They are not duplicates and should not be clustered as the same underlying change.", + "summary": "These are two unrelated feature pull requests in different model families: TimesFM covariate forecasting vs. DINOv3 image classification support. They should not be merged or triaged as duplicates.", "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "No canonical PR as a duplicate target: the PRs address different models, code paths, and user-facing capabilities. If one must be chosen as the representative, 41224 is the more narrowly scoped and self-contained change.", "best_issue_reason": null, - "best_pr_reason": "No true duplicate exists here, but PR 41224 is the cleaner representative because it is narrowly scoped to one model feature and has a more self-contained change set than the broader TimesFM covariate PR.", + "best_pr_reason": "pull_request:41224 is the best single PR in the cluster because it is focused on one concrete capability addition with a tight implementation surface, while 40755 is a separate larger feature for a different model.", "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "Different models, different functionality, and different affected files/code paths. TimesFM adds covariate-based forecasting; DINOv3 adds image classification support. They are unrelated changes and not mergeable into one PR." + "reason": "Different underlying changes: TimesFM covariate forecasting vs. DINOv3 image classification support. They do not fix the same code-path problem and could not plausibly be merged into one PR." } ] }, @@ -18013,7 +17911,7 @@ { "label": "pull_request", "component_index": 4, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18076,7 +17974,7 @@ { "label": "pull_request", "component_index": 5, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18094,7 +17992,7 @@ "pull_request:41797|pull_request:45186", "pull_request:45186|pull_request:45630" ], - "prepared_review_unit_hash": "2708cfeb1d55b76c206825a3f1f4b86108e260d9d29e8c8a820b2aae4db943e0", + "prepared_review_unit_hash": "fa6d7c238f155cb13c29b120743fdb2d0a50caf556bba95fe7a5d8a365bd5947", "component_budget": { "node_count": 3, "item_count": 3, @@ -18116,24 +18014,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition PRs (DeepSeek OCR, Isaac, Kimi2-6). They share framework scaffolding changes, but they do not fix the same underlying code-path problem, so no duplicate edges should be accepted.", - "confidence": 0.98, + "summary": "These are three separate model-integration pull requests (DeepSeek OCR, Isaac, Kimi2-6), not duplicate fixes for the same underlying bug or change. The shared files are just common Transformers boilerplate for new model additions, so the soft links should be rejected.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45630 is the most mature and active of the three, with the broadest model-integration footprint, so it is the best single representative if one must be chosen. It is still not a duplicate of the others.", + "canonical_pr_reason": "No true canonical duplicate exists here because the PRs add different models and implement different code paths. If forced to pick one representative, 45630 is the strongest candidate because it has the broadest implementation surface and highest review activity.", "best_issue_reason": null, - "best_pr_reason": "PR 45630 has the strongest standalone signal as a completed model integration effort, with the highest review activity and a substantial end-to-end implementation across autos, processors, and tests.", + "best_pr_reason": "45630 is the most substantial and reviewed of the three new-model PRs, with extensive auto-mapping and model files plus test coverage, so it is the best representative of this cluster even though it is not semantically the same change as the others.", "soft_edge_verdicts": [ { "left": "pull_request:41797", "right": "pull_request:45186", "accept": false, - "reason": "Both are new-model integration PRs, but DeepSeek OCR and Isaac are different models with different codepaths and files; shared docs/auto-registration scaffolding is not enough to treat them as the same change." + "reason": "Different model additions: DeepSeek OCR vs Isaac. Shared Transformers scaffolding files do not indicate the same bug or change, and these would not be merged as one PR." }, { "left": "pull_request:45186", "right": "pull_request:45630", "accept": false, - "reason": "Isaac and Kimi2-6 are distinct model additions touching different model families and implementations. Similar boilerplate around auto mappings/docs does not indicate one underlying bug or mergeable combined PR." + "reason": "Different model additions: Isaac vs Kimi2-6. The overlap is generic integration plumbing, not a shared concrete code-path fix or feature." } ] }, @@ -18147,7 +18045,7 @@ { "label": "pull_request", "component_index": 6, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18163,12 +18061,12 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "f6f1ad58e018d0c13e0106dbac712d77d1c887e4329b80733021f58e50761601", + "prepared_review_unit_hash": "019d8a8ef99a0fa869b71af98f60cf0bb7d5c71cae52ed2ae02edc2a4b18f90a", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5935, + "serialized_chars": 5934, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -18176,7 +18074,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5935, + "serialized_chars": 5934, "estimated_input_tokens": 1484, "estimated_eval_tokens": 3224 }, @@ -18185,18 +18083,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related workflow dependency upgrades, but they target different GitHub Actions families and different code paths: one is a broad Node 24 compatibility bump for core actions, the other is a docker/build-push-action upgrade. They are not duplicates.", - "confidence": 0.95, + "summary": "These are two separate GitHub Actions maintenance PRs, not the same underlying change: one is a broad Node 24 compatibility upgrade for workflow actions, while the other mainly bumps docker/build-push-action and related workflow action versions. The overlap in workflow files is incidental.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 42975 is the better canonical representative because it has the clearest stated purpose (Node 24 compatibility) and the broader workflow-wide actions upgrade scope.", + "canonical_pr_reason": "No clear canonical duplicate PR: the two PRs address different dependency upgrade goals and should remain separate.", "best_issue_reason": null, - "best_pr_reason": "PR 42975 best captures the cluster\u2019s general theme of GitHub Actions maintenance across workflows, while remaining more explicit and broader than the docker-only bump in 42976.", + "best_pr_reason": "PR 42975 is the broader, more central workflow-actions upgrade and is the better representative of the cluster theme, but it is still not a duplicate of 42976.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Both are workflow dependency upgrades, but they change different action packages and likely different runtime paths: core actions versions in 42975 versus docker/build-push-action in 42976. They could coexist in one maintenance stream, but they are not the same underlying change." + "reason": "Different concrete changes: 42975 targets Node 24 compatibility across GitHub Actions, while 42976 upgrades docker/build-push-action and a smaller set of related workflow actions. Shared workflow paths alone are not enough to treat them as the same PR." } ] }, @@ -18210,7 +18108,7 @@ { "label": "pull_request", "component_index": 7, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18218,13 +18116,18 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-42978-5", + "cluster_id": "cluster-42978-10", "nodes": [ "pull_request:42978", "pull_request:43448", "pull_request:43451", + "pull_request:45487", "pull_request:45490", - "pull_request:46162" + "pull_request:46109", + "pull_request:46135", + "pull_request:46151", + "pull_request:46162", + "pull_request:46214" ], "soft_pairs": [ "pull_request:42978|pull_request:43448", @@ -18233,78 +18136,134 @@ "pull_request:43448|pull_request:43451", "pull_request:43448|pull_request:45490", "pull_request:43451|pull_request:45490", - "pull_request:45490|pull_request:46162" + "pull_request:45487|pull_request:46214", + "pull_request:45487|pull_request:46151", + "pull_request:45490|pull_request:46162", + "pull_request:46109|pull_request:46214", + "pull_request:46109|pull_request:46151", + "pull_request:46135|pull_request:46214", + "pull_request:46135|pull_request:46151", + "pull_request:46151|pull_request:46214", + "pull_request:46162|pull_request:46214" ], - "prepared_review_unit_hash": "f37b0e4482342b26bfd2d03b659f14f0b9319c01b52abedce605031e2da4e3ed", + "prepared_review_unit_hash": "3a023daa2042dcf1a6d6de47911619e8766eeb251314b097fa39e41c028b1b6d", "component_budget": { - "node_count": 5, - "item_count": 5, - "soft_pair_count": 7, - "serialized_chars": 16552, - "estimated_input_tokens": 4138, - "estimated_eval_tokens": 8532 + "node_count": 10, + "item_count": 10, + "soft_pair_count": 15, + "serialized_chars": 29975, + "estimated_input_tokens": 7494, + "estimated_eval_tokens": 15244 }, "budget": { - "node_count": 5, - "item_count": 5, - "soft_pair_count": 7, - "serialized_chars": 16552, - "estimated_input_tokens": 4138, - "estimated_eval_tokens": 8532 + "node_count": 10, + "item_count": 10, + "soft_pair_count": 15, + "serialized_chars": 29975, + "estimated_input_tokens": 7494, + "estimated_eval_tokens": 15244 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are mostly unrelated model-addition efforts plus one small test-only change. The shared files are generic integration points (docs/auto registries/tests), not evidence of the same underlying bug or feature.", - "confidence": 0.96, + "summary": "These PRs form two loose themes\u2014new model additions and XPU/test expectation tweaks\u2014but none of the soft edges look like true duplicates of the same concrete change. The similarities are mostly broad-file overlap (docs/auto mappings/tests) rather than shared bug fixes.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43451 (Molmo2) is the strongest standalone representative: it has the largest scope, highest discussion activity, and a complete model addition across code, docs, and tests. The others are different model additions or a test tweak.", + "canonical_pr_reason": "No single canonical PR stands out: the set mixes unrelated model additions (Molmo, Molmo2, CTSM, ViT NEPA) with unrelated test-only XPU/distributed updates, so there is no one artifact that safely represents the cluster.", "best_issue_reason": null, - "best_pr_reason": "#43451 is the best PR in the set because it is the most substantial and self-contained change, but it is not a duplicate of the other PRs.", + "best_pr_reason": "No clear global best PR for duplicate-triage purposes; the apparent overlaps are superficial and the PRs address different models or different test paths.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo. Shared registry/docs files are expected boilerplate, not the same code-path change." + "reason": "Both are model-addition PRs, but for different model families (ViT NEPA vs Molmo); shared docs/auto files are just repository plumbing." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "Different model additions: ViT NEPA vs Molmo2. They touch common auto-registry files but implement separate models and behaviors." + "reason": "Different model additions (ViT NEPA vs Molmo2); not the same underlying change." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "Different model additions in different domains: vision model ViT NEPA vs time-series CTSM. No shared underlying bug or fix." + "reason": "Unrelated model additions (ViT NEPA vs CTSM); overlap is only in shared registration/docs files." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are related by name only; they are separate model integrations with different files, tests, and processing code." + "reason": "Molmo and Molmo2 are distinct model additions, not one duplicate change." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo is a vision-language model addition, while CTSM is a time-series model addition. Shared infra files do not indicate the same change." + "reason": "Molmo vs CTSM are unrelated model adds; shared auto/docs edits are incidental." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 and CTSM are distinct model additions with different architectures and code paths; not mergeable into one PR as the same fix." + "reason": "Molmo2 and CTSM are different models with different codepaths and tests." + }, + { + "left": "pull_request:45487", + "right": "pull_request:46214", + "accept": false, + "reason": "AltCLIP/ChineseCLIP model-parallel fix is unrelated to HRM text XPU expectation updates." + }, + { + "left": "pull_request:45487", + "right": "pull_request:46151", + "accept": false, + "reason": "Different problems: model-parallel device handling vs pipeline token-classification test expectations." }, { "left": "pull_request:45490", "right": "pull_request:46162", "accept": false, - "reason": "CTSM model integration is unrelated to the XPU determinism test annotations in existing Florence2/InstructBLIP/LFM2-VL tests. One adds a model; the other adjusts test stability." + "reason": "CTSM model implementation PR vs Florence2 XPU test adjustments; not the same bug or change." + }, + { + "left": "pull_request:46109", + "right": "pull_request:46214", + "accept": false, + "reason": "Both touch tests, but one is distributed trainer parity and the other is HRM text expectations; different concrete test paths." + }, + { + "left": "pull_request:46109", + "right": "pull_request:46151", + "accept": false, + "reason": "Distributed trainer test genericization and token-classification expectations are separate changes." + }, + { + "left": "pull_request:46135", + "right": "pull_request:46214", + "accept": false, + "reason": "Kosmos2.5 XPU expectations and HRM text XPU expectations are model-specific, not a single duplicate fix." + }, + { + "left": "pull_request:46135", + "right": "pull_request:46151", + "accept": false, + "reason": "Different model test suites and different expectation updates; not mergeable as one PR." + }, + { + "left": "pull_request:46151", + "right": "pull_request:46214", + "accept": false, + "reason": "Token-classification pipeline expectations vs HRM text expectations are unrelated model/test updates." + }, + { + "left": "pull_request:46162", + "right": "pull_request:46214", + "accept": false, + "reason": "Florence2 XPU expectations and HRM text XPU expectations are distinct model-specific test changes." } ] }, @@ -18318,7 +18277,7 @@ { "label": "pull_request", "component_index": 8, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18339,7 +18298,7 @@ "pull_request:43665|pull_request:46102", "pull_request:43665|pull_request:44815" ], - "prepared_review_unit_hash": "b71a89102b11030e8da83e962addb10105d64397c4ce474ff94fbf363e09ee90", + "prepared_review_unit_hash": "21c978c73e3a32f2fb1de0c1ae4a2239380534c24e5e4aa1a22074200ee57a22", "component_budget": { "node_count": 4, "item_count": 4, @@ -18361,36 +18320,36 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are four unrelated PRs across whisper decoding, model test tweaks, FP8 dequantization, and FSDP tied-weights cleanup. The only nontrivial code-fix PR is the FP8 dequant change; the others are either test-only or target different subsystems, so no soft duplicates should be merged.", - "confidence": 0.93, + "summary": "These PRs are not duplicates: two are test-only edits, one fixes Whisper decoding, one fixes dequant/FP8 loading, and one fixes FSDP tied-weights handling. They touch different code paths and should stay separate.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44815 is the most concrete fix: it changes core loading and FP8 integration code with a clear dequant/padding path update, backed by tests, and represents a substantive bug fix rather than a superficial test edit.", + "canonical_pr_reason": "If a representative PR is needed, 44815 is the strongest standalone code fix: it changes real loading/quantization logic across multiple files, while the others are either test-only or different subsystems.", "best_issue_reason": null, - "best_pr_reason": "PR 44815 is the best canonical PR because it addresses a specific runtime code path with clear implementation changes and test coverage; the other PRs are either unrelated or look like placeholder/test-only edits.", + "best_pr_reason": "44815 is the best representative of a substantive bug fix in this cluster; it has concrete implementation changes, unlike 43665 which is test-only and the others which address unrelated areas.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Both are mostly test-related, but they target different models and different code paths; no evidence they fix the same underlying bug." + "reason": "Whisper tokenizer decoding vs CLIP/ViT test edits; different subsystems and no shared bug/change." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper tokenizer decoding and FP8 dequantization are unrelated subsystems with no shared bug or change surface." + "reason": "Whisper timestamp/batch decode fix is unrelated to FP8 dequant/loading changes." }, { "left": "pull_request:43665", "right": "pull_request:46102", "accept": false, - "reason": "A pair of test edits in CLIP/ViT versus an FSDP tied-weights fix are clearly unrelated." + "reason": "Test-only CLIP/ViT assertions are unrelated to FSDP tied-weights logic." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "Despite generic similarity, these touch different modules and different problems; they would not plausibly be merged as one PR." + "reason": "Both touch tests superficially, but the underlying changes are unrelated: model tests vs quantization/loading code." } ] }, @@ -18404,7 +18363,7 @@ { "label": "pull_request", "component_index": 9, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18427,7 +18386,7 @@ "pull_request:45932|pull_request:46100", "pull_request:46100|pull_request:46114" ], - "prepared_review_unit_hash": "b54e08caee55f0ebfff5d265fa8a15e514c904562d3e80558d916b16de554891", + "prepared_review_unit_hash": "9659f88bb7446b99f073d17e482942180306a7fa696ed903be67ac33d0421692", "component_budget": { "node_count": 5, "item_count": 5, @@ -18449,48 +18408,48 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Only one soft pair looks like a true duplicate: the two forked test-migration PRs with identical viT test edits. The other pairs share a file or theme but change different code paths and should stay separate.", - "confidence": 0.89, + "summary": "Only 46100 and 46114 are true duplicates: they carry the same one-file ViT test migration diff. The other PRs are test/noise PRs or broader unrelated changes that merely overlap on a filename or subsystem.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #46100 is the cleanest representative of the exact duplicate pair: #46114 has the same diff in the same test file and appears to be a forked copy.", + "canonical_pr_reason": "PR 46100 is the clean canonical representative: it matches 46114 exactly at the diff level, and its title is the simpler of the duplicate pair.", "best_issue_reason": null, - "best_pr_reason": "PR #46100 is the best overall canonical PR because it is the clearest standalone instance of the duplicated test-migration change.", + "best_pr_reason": "46100 is the best overall PR candidate because it is the clearest exact duplicate target in the cluster and represents the only truly mergeable duplicate pair.", "soft_edge_verdicts": [ { "left": "pull_request:43488", - "right": "pull_request:43532", + "right": "pull_request:46114", "accept": false, - "reason": "Both are unrelated fake/test PRs and only overlap on a shared ViT test file; the actual edits are different (setup/dependency changes vs workflow/other model files)." + "reason": "Shared ViT test filename is incidental; 43488 changes setup/version/dependencies while 46114 only edits test assertions. Not the same underlying change." }, { "left": "pull_request:43488", - "right": "pull_request:46114", + "right": "pull_request:43532", "accept": false, - "reason": "They both touch the ViT test file, but the surrounding changes and intent are different, so they are not the same underlying change." + "reason": "These are different fake PRs with different code paths and files; the overlap is too broad and not a concrete duplicate." }, { "left": "pull_request:43532", "right": "pull_request:46114", "accept": false, - "reason": "Shared ViT test-file overlap is too broad; #43532 also modifies workflows and other model files, so this is not one mergeable change." + "reason": "43532 is a workflow/infra change, while 46114 is a local test-file edit. They do not fix the same bug or implement the same change." }, { "left": "pull_request:45932", "right": "pull_request:46100", "accept": false, - "reason": "These are different kinds of PRs: #45932 changes model code/workflows across several files, while #46100 only adds failing tests in ViT." + "reason": "45932 is a broad multi-file model/config/workflow change; 46100 is a narrow ViT test migration. Same broad area, but not the same underlying code change." }, { "left": "pull_request:46100", "right": "pull_request:46114", "accept": true, - "reason": "The diffs are effectively identical in the same file, and #46114 is described as a forked test-migration PR, so they look like the same underlying change." + "reason": "Exact same diff in tests/models/vit/test_modeling_vit.py; these are the same test migration change and could plausibly be merged as one PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted duplicate pair, 46100/46114, has identical diffs in the same test file; the other pairs share at most a file or broad theme but include different surrounding changes, so the false verdicts are reasonable." + "feedback": "The summary is grounded in the packet. The only accepted soft edge, 46100/46114, is supported by identical diff previews in the same file, and the rejected pairs are conservative given the broader or unrelated changes. Canonical PR 46100 is a reasonable duplicate representative." }, "evaluator_used": true, "retried": false, @@ -18501,7 +18460,7 @@ { "label": "pull_request", "component_index": 10, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18517,7 +18476,7 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "491b9e094c8da258bc44b1e2f2d0801e0aa951a3c924ab8270c415facfdf7b1c", + "prepared_review_unit_hash": "5a07b8df0a7e47e3c115639cd3ee798ee1184dd2927552ec18ce7b21f4180cc8", "component_budget": { "node_count": 2, "item_count": 2, @@ -18539,18 +18498,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related but distinct PRs: one adds a promptable concept segmentation pipeline, the other adds a promptable visual segmentation pipeline with different supporting models and tests. They share docs/registry plumbing, but they do not fix the same concrete code path and should not be merged as duplicates.", + "summary": "These PRs are closely related in theme and documentation scaffolding, but they implement different pipelines for different segmentation tasks and model integrations. They should not be treated as duplicates.", "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical duplicate PR: the two changes target different segmentation tasks and different model integrations, so there is no single PR that cleanly subsumes the other as the same change.", + "canonical_pr_reason": "No canonical PR: 43612 adds Promptable Concept Segmentation (SAM3-focused), while 43613 adds Promptable Visual Segmentation with different model hooks and test coverage.", "best_issue_reason": null, - "best_pr_reason": "If a representative PR is needed, #43613 is the broader and more central pipeline addition, touching more model backends and test coverage; but it is not a duplicate of #43612.", + "best_pr_reason": "43613 is the better representative PR for the cluster because it is the broader, more fully integrated pipeline addition; 43612 is a separate narrower feature.", "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Same broad domain and shared documentation/registry edits, but they implement different pipelines for different segmentation tasks with different filenames, backends, and tests. Not the same underlying bug/change." + "reason": "Reject: they share pipeline/docs boilerplate, but the underlying changes are different features with different code paths and model/test integrations (concept segmentation vs visual segmentation)." } ] }, @@ -18564,7 +18523,7 @@ { "label": "pull_request", "component_index": 11, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18580,40 +18539,40 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "b637c58bd7796baa716212ae2db677532a04259ec6426ddcee6c7652c4cf34c3", + "prepared_review_unit_hash": "17aca8c302f0419d5d93e8cfcb5c87d1718c26f9a34f6322cf81a2d93b29e603", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4293, - "estimated_input_tokens": 1074, - "estimated_eval_tokens": 2404 + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4293, - "estimated_input_tokens": 1074, - "estimated_eval_tokens": 2404 + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two PRs are unrelated Trainer enhancements that happen to touch the same file: one changes gradient-accumulation loss scaling behavior, the other extracts signature-column computation for column filtering/PEFT overrides. They are not the same underlying change.", - "confidence": 0.98, + "summary": "These are two unrelated trainer.py pull requests: one adds `_loss_is_scaled_for_ga` for gradient-accumulation loss scaling, the other extracts `_get_signature_columns` for column filtering. Same file, different code paths, not the same underlying change.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: 43651 and 43663 address different code paths and different extension points in trainer.py, so neither should subsume the other.", + "canonical_pr_reason": "No canonical PR: the patches solve different Trainer extension-point problems and are not mergeable into one coherent fix.", "best_issue_reason": null, - "best_pr_reason": "No single best PR for duplication purposes; these are separate changes and should remain independently triaged rather than merged as one representative.", + "best_pr_reason": "Neither PR is a good global representative for the cluster because they address unrelated behaviors in `Trainer` despite touching the same file.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Reject: one PR adds _loss_is_scaled_for_ga to control gradient-accumulation loss normalization, while the other adds _get_signature_columns for trainer input filtering. Same file, but different behavior and unrelated bugs/features." + "reason": "Different functionality and different code paths: one changes gradient-accumulation loss normalization, the other refactors signature-column inspection for data filtering." } ] }, @@ -18627,7 +18586,7 @@ { "label": "pull_request", "component_index": 12, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18643,12 +18602,12 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "c9c2fdd6902b9a274c0c0f58d794652df23f0eeb9db6a0c3e3c848651d4940af", + "prepared_review_unit_hash": "f60eca232ef0b4e630eb6a8a5365ab10bae824d47494678849c386f5b5098c28", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3662, + "serialized_chars": 3664, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -18656,7 +18615,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3662, + "serialized_chars": 3664, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -18665,18 +18624,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs touch the same CLI file and both mention TypeAdapter/name-resolution issues, but they appear to fix different root causes: one is a broader typing/postponed-annotation cleanup, the other is a pydantic-not-installed import/type-checking fix.", + "summary": "Two open PRs touch the same file and both address annotation-related NameErrors in `src/transformers/cli/serve.py`, but they fix different runtime paths: one is a broad typing/future-annotations cleanup, the other is specifically about `TypeAdapter` when `pydantic` is absent.", "confidence": 0.87, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #43656 is the broader, earlier CLI typing fix and is the better representative of the cluster\u2019s shared TypeAdapter/name-resolution theme.", + "canonical_pr_reason": "PR #43656 is the broader representative fix for import-time typing/annotation errors in the CLI, with more review/discussion and a wider cleanup than the narrower `TypeAdapter`-specific patch.", "best_issue_reason": null, - "best_pr_reason": "PR #43842 is the more concrete fix for a specific runtime failure mode in `serve.py` (avoiding `TypeAdapter` import/name errors when pydantic is absent).", + "best_pr_reason": "PR #43656 most centrally captures the cluster\u2019s underlying serve.py annotation problem and is the more comprehensive change set.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", "accept": false, - "reason": "Same file and similar symptom class, but different underlying bugs and fixes: one adds postponed annotations and typing cleanup, the other makes `TypeAdapter` a TYPE_CHECKING-only reference for the no-pydantic case. They would not naturally merge as one PR." + "reason": "Related but not duplicates: #43656 adds `from __future__ import annotations` and broad type-hint cleanup, while #43842 specifically guards/quotes `TypeAdapter` for the no-pydantic case. They address different concrete NameError paths and could be merged separately." } ] }, @@ -18690,7 +18649,7 @@ { "label": "pull_request", "component_index": 13, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -18756,7 +18715,7 @@ { "label": "pull_request", "component_index": 14, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 2, "status": "reviewed", @@ -18805,16 +18764,16 @@ "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:44066|pull_request:44068", "pull_request:44019|pull_request:44722", "pull_request:44086|pull_request:44722", "pull_request:44019|pull_request:44085", "pull_request:44018|pull_request:44019", "pull_request:44019|pull_request:44071", "pull_request:44068|pull_request:44116", - "pull_request:44068|pull_request:44722", "pull_request:44018|pull_request:44116", + "pull_request:44068|pull_request:44722", "pull_request:44071|pull_request:44722", "pull_request:44018|pull_request:44086", "pull_request:44018|pull_request:44085", @@ -18829,33 +18788,33 @@ "pull_request:44017|pull_request:44161", "pull_request:44017|pull_request:44073", "pull_request:44068|pull_request:44071", - "pull_request:43997|pull_request:44073", "pull_request:44071|pull_request:44086", - "pull_request:44073|pull_request:44101", + "pull_request:43997|pull_request:44073", "pull_request:44017|pull_request:44071", - "pull_request:44056|pull_request:44101", - "pull_request:44073|pull_request:44161", + "pull_request:44073|pull_request:44101", "pull_request:44059|pull_request:44071", - "pull_request:44059|pull_request:44068", "pull_request:44018|pull_request:44059", + "pull_request:44073|pull_request:44161", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44068", "pull_request:44101|pull_request:44161", - "pull_request:43997|pull_request:44017", + "pull_request:44017|pull_request:44018", "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:44056|pull_request:44161", "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:44010|pull_request:44073", "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:44071|pull_request:44129", + "pull_request:43997|pull_request:44129", "pull_request:43996|pull_request:44161", "pull_request:43996|pull_request:44017", "pull_request:44010|pull_request:44044", @@ -18868,100 +18827,112 @@ "pull_request:44044|pull_request:44129", "pull_request:44044|pull_request:44154" ], - "prepared_review_unit_hash": "fd74af16b5cea9a3db0680b6fd326c5a5abbc4e374ff86ad2fa87b0dd7962c60", + "prepared_review_unit_hash": "274d4d38f118395514d4dbd5706eeed927fc123e9abf8e28d63166c45d66481d", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84894, - "estimated_input_tokens": 21224, - "estimated_eval_tokens": 42704 + "serialized_chars": 84899, + "estimated_input_tokens": 21225, + "estimated_eval_tokens": 42706 }, "budget": { "node_count": 27, "item_count": 27, "soft_pair_count": 72, - "serialized_chars": 75091, - "estimated_input_tokens": 18773, - "estimated_eval_tokens": 37802 + "serialized_chars": 75096, + "estimated_input_tokens": 18774, + "estimated_eval_tokens": 37804 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of independent model-specific refactors to standardize output tracing/capture decorators, often all tied to issue #43979. They share a theme but are not one duplicate bug fix across the whole set; only a few GPT-Neo/GPT-J overlaps look like true same-change duplicates.", + "summary": "This cluster is mostly a wave of standardized output-tracing refactors across many model backbones. Many of the relevant PRs point to #43979, but several do not, so it\u2019s better framed as a broad theme plus a GPT-style subset where same-file duplicates are plausible.", "confidence": 0.87, - "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44722 is the best representative of the overlapping subgroup: it is the broadest and clearest standardized output-tracing refactor, covering the same GPT-J change as earlier PRs and also extending the pattern to CodeGen.", - "best_issue_reason": null, - "best_pr_reason": "44722 is the most suitable global representative because it is the most complete later version of the shared GPT-J/CodeGen output-tracing refactor; the other PRs are separate model-specific applications of the same framework idea.", + "canonical_issue_reason": "#43979 is the shared umbrella tracker for the output-tracing modernization work in the relevant PRs, though not every PR in the cluster explicitly links it.", + "canonical_pr_reason": "PR 44722 is the clearest representative because it explicitly targets #43979 and captures the core GPT-J/CodeGen output-tracing refactor in the same code path as the GPT-style duplicates.", + "best_issue_reason": "#43979 is the best cluster anchor because it ties together the main standardized output-tracing refactor wave, especially the GPT-style subset.", + "best_pr_reason": "PR 44722 is the best single representative: it is explicit about the shared tracker, spans the core GPT-J/CodeGen implementation, and best summarizes the duplicate-prone subset.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Different models and code paths (CVT/FNet vs GPT-J); same tracking issue is not enough to make them duplicates." + "reason": "Different model families and code paths; shared output-tracing theme is not enough." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Both are output-tracing refactors, but for different architectures (CVT/FNet vs DeBERTa v2) and different implementation details." + "reason": "Different backbones and refactor scopes; same tracker/theme alone does not make them duplicates." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Both modify GPT-J output tracing in the same file with the same decorator-style refactor; 44066 is the broader sibling change with the same underlying GPT-J fix." + "reason": "Same GPT-J output-tracing code path in the same file; this looks like an overlapping slice of one refactor." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "ResNet/RegNet vs EfficientNet are separate model-specific refactors, not the same concrete change." + "reason": "ResNet/RT-DETR ResNet versus EfficientNet are different model implementations, so these are separate changes." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "EfficientNet output tracing is unrelated to GPT-J/CodeGen; only the high-level theme overlaps." + "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths; only the refactor pattern matches." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "GPT-J/CodeGen versus EfficientNet are different code paths and different PRs." + "reason": "GPT-J/CodeGen versus EfficientNet are different implementations and should not be merged as one PR." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "MobileNetV2 and DeBERTa v2 are distinct implementations; this is only a shared refactor pattern." + "reason": "MobileNetV2 and DeBERTa are unrelated model-specific refactors." }, { - "left": "pull_request:44018", - "right": "pull_request:44068", - "accept": true, - "reason": "Both target GPT-Neo output tracing and appear to implement the same decorator-based migration in the same model family." + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen and MGP-STR are different model families and code paths." }, { "left": "pull_request:44066", - "right": "pull_request:44068", + "right": "pull_request:44071", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are different models and different code paths." + "reason": "GPT-J/CodeGen versus MPT are distinct implementations; the shared decorator pattern is not enough." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Same GPT-Neo output-tracing refactor in the same model/file; these plausibly duplicate the same change." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "GPT-Neo vs GPT-J/CodeGen are different model refactors, even though they share the broader output-tracing theme." + "reason": "GPT-Neo and GPT-J/CodeGen are different model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen and GPT-Neo are different code paths, so these are not duplicates." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet output tracing refactor is unrelated to GPT-J/CodeGen." + "reason": "ResNet and GPT-J/CodeGen are unrelated backbones despite similar refactor wording." }, { "left": "pull_request:44086", @@ -18973,364 +18944,178 @@ "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "ResNet and GPT-J are different models and different concrete fixes." + "reason": "ResNet and GPT-J are unrelated implementations; same theme is not enough." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet refactors are unrelated beyond the generic output-tracing theme." - }, - { - "left": "pull_request:44019", - "right": "pull_request:44071", - "accept": false, - "reason": "ResNet and MPT are distinct model implementations with separate changes." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44116", - "accept": false, - "reason": "GPT-Neo and Flaubert are different model code paths." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44722", - "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are different model-specific refactors." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44116", - "accept": false, - "reason": "GPT-Neo vs Flaubert are different models and not the same change." - }, - { - "left": "pull_request:44017", - "right": "pull_request:44722", - "accept": false, - "reason": "SegFormer and GPT-J/CodeGen are separate model refactors." - }, - { - "left": "pull_request:44017", - "right": "pull_request:44161", - "accept": false, - "reason": "SegFormer and LongT5 are unrelated model implementations." - }, - { - "left": "pull_request:44017", - "right": "pull_request:44073", - "accept": false, - "reason": "SegFormer and VisualBERT are different models; same issue target is insufficient." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44071", - "accept": false, - "reason": "GPT-Neo and MPT are separate model changes." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44086", - "accept": false, - "reason": "GPT-Neo and MGP-STR are distinct model-specific refactors." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44086", - "accept": false, - "reason": "GPT-Neo vs MGP-STR are different code paths and fixes." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44071", - "accept": false, - "reason": "GPT-Neo and MPT are separate model implementations." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44073", - "accept": false, - "reason": "RegNet and VisualBERT are unrelated model refactors." - }, - { - "left": "pull_request:44071", - "right": "pull_request:44086", - "accept": false, - "reason": "MPT and MGP-STR are separate model-specific output changes." - }, - { - "left": "pull_request:44073", - "right": "pull_request:44101", - "accept": false, - "reason": "VisualBERT and Flaubert/XLM are different model families." - }, - { - "left": "pull_request:44017", - "right": "pull_request:44071", - "accept": false, - "reason": "SegFormer and MPT are unrelated changes." - }, - { - "left": "pull_request:44056", - "right": "pull_request:44101", - "accept": false, - "reason": "MPNet and Flaubert/XLM are different models and fixes." - }, - { - "left": "pull_request:44073", - "right": "pull_request:44161", - "accept": false, - "reason": "VisualBERT and LongT5 are unrelated model-specific changes." - }, - { - "left": "pull_request:44059", - "right": "pull_request:44071", - "accept": false, - "reason": "GPT-2 and MPT are separate model refactors." - }, - { - "left": "pull_request:44059", - "right": "pull_request:44068", - "accept": false, - "reason": "GPT-2 and GPT-Neo are different models with different output paths." - }, - { - "left": "pull_request:44018", - "right": "pull_request:44059", - "accept": false, - "reason": "GPT-Neo and GPT-2 are distinct model implementations." - }, - { - "left": "pull_request:44101", - "right": "pull_request:44161", - "accept": false, - "reason": "Flaubert/XLM and LongT5 are different model families." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44017", - "accept": false, - "reason": "RegNet and SegFormer are different vision models; same issue target is only thematic." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44056", - "accept": false, - "reason": "RegNet and MPNet are unrelated model-specific refactors." - }, - { - "left": "pull_request:44071", - "right": "pull_request:44129", - "accept": false, - "reason": "MPT and SpeechT5 are different model code paths." + "reason": "GPT-Neo and ResNet are different model families and not the same concrete change." }, { - "left": "pull_request:44074", - "right": "pull_request:44101", + "left": "pull_request:44019", + "right": "pull_request:44071", "accept": false, - "reason": "TextNet and Flaubert/XLM are unrelated model implementations." + "reason": "ResNet and MPT are separate model-specific refactors." }, { - "left": "pull_request:44056", - "right": "pull_request:44161", + "left": "pull_request:44068", + "right": "pull_request:44086", "accept": false, - "reason": "MPNet and LongT5 are separate refactors for different architectures." + "reason": "GPT-Neo and MGP-STR are unrelated model implementations." }, { - "left": "pull_request:43997", - "right": "pull_request:44018", + "left": "pull_request:44018", + "right": "pull_request:44116", "accept": false, - "reason": "RegNet and GPT-Neo are unrelated model-specific changes." + "reason": "GPT-Neo and Flaubert/XLM are different code paths; only the refactor style is shared." }, { - "left": "pull_request:43997", - "right": "pull_request:44161", + "left": "pull_request:44068", + "right": "pull_request:44722", "accept": false, - "reason": "RegNet and LongT5 are different model families." + "reason": "GPT-Neo versus GPT-J/CodeGen are different implementations, so this is not a duplicate." }, { - "left": "pull_request:44074", - "right": "pull_request:44161", + "left": "pull_request:44071", + "right": "pull_request:44722", "accept": false, - "reason": "TextNet and LongT5 do not share the same concrete change." + "reason": "MPT and GPT-J/CodeGen are unrelated model families." }, { "left": "pull_request:44018", - "right": "pull_request:44161", - "accept": false, - "reason": "GPT-Neo and LongT5 are distinct model changes." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44071", + "right": "pull_request:44059", "accept": false, - "reason": "RegNet and MPT are unrelated beyond the shared refactor theme." + "reason": "GPT-Neo and GPT-2 are separate model implementations and should stay separate." }, { - "left": "pull_request:44010", - "right": "pull_request:44071", + "left": "pull_request:44085", + "right": "pull_request:44086", "accept": false, - "reason": "SqueezeBERT and MPT are different model-specific refactors." + "reason": "GPT-J and MGP-STR are unrelated apart from the shared output-tracing pattern." }, { - "left": "pull_request:44010", - "right": "pull_request:44073", + "left": "pull_request:44098", + "right": "pull_request:44101", "accept": false, - "reason": "SqueezeBERT and VisualBERT are separate model code paths." + "reason": "ViLT and Flaubert/XLM are different model families with distinct code paths." }, { - "left": "pull_request:44018", - "right": "pull_request:44129", + "left": "pull_request:44017", + "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo and SpeechT5 are different model families." + "reason": "SegFormer and GPT-J/CodeGen are unrelated model implementations." }, { - "left": "pull_request:43996", + "left": "pull_request:44017", "right": "pull_request:44161", "accept": false, - "reason": "CVT/FNet and LongT5 are unrelated model-specific changes." + "reason": "SegFormer and LongT5 are different architectures; only the refactor style matches." }, { - "left": "pull_request:43996", - "right": "pull_request:44017", + "left": "pull_request:44017", + "right": "pull_request:44073", "accept": false, - "reason": "CVT/FNet and SegFormer are separate vision model implementations with different changes." + "reason": "SegFormer and VisualBERT are different code paths and not duplicates." }, { - "left": "pull_request:44010", - "right": "pull_request:44044", + "left": "pull_request:44068", + "right": "pull_request:44071", "accept": false, - "reason": "SqueezeBERT and DeBERTa v2 are different models and code paths." + "reason": "GPT-Neo and MPT are different model-specific output-tracing changes." }, { - "left": "pull_request:44044", - "right": "pull_request:44074", + "left": "pull_request:44059", + "right": "pull_request:44086", "accept": false, - "reason": "DeBERTa v2 and TextNet are unrelated model-specific changes." - }, - { - "left": "pull_request:44066", - "right": "pull_request:44722", - "accept": true, - "reason": "Same GPT-J output-tracing refactor; 44722 is the broader later version that includes the same GPT-J fix." + "reason": "GPT-2 and MGP-STR are separate model implementations." }, { "left": "pull_request:44085", "right": "pull_request:44722", "accept": true, - "reason": "Both include the same GPT-J decorator-based output-tracing change in the same file, so they look like the same underlying fix." + "reason": "Same GPT-J file and same decorator-based refactor; 44722 just broadens coverage with the same underlying GPT-J change." }, { "left": "pull_request:44085", "right": "pull_request:44129", "accept": false, - "reason": "GPT-J and SpeechT5 are unrelated model-specific changes." + "reason": "GPT-J and SpeechT5 are unrelated code paths; the shared tracker does not make them duplicates." }, { "left": "pull_request:43996", "right": "pull_request:44007", "accept": false, - "reason": "CVT/FNet and RegNet/ResNet/RT-DETR are different model families; same issue target is not enough." + "reason": "Different model families and different forward/output plumbing changes." }, { "left": "pull_request:43996", "right": "pull_request:44129", "accept": false, - "reason": "CVT/FNet and SpeechT5 are unrelated code paths." - }, - { - "left": "pull_request:44066", - "right": "pull_request:44086", - "accept": false, - "reason": "GPT-J/CodeGen and MGP-STR are unrelated model-specific refactors." + "reason": "CVT/FNet and SpeechT5 are unrelated implementations." }, { - "left": "pull_request:44066", - "right": "pull_request:44071", + "left": "pull_request:44044", + "right": "pull_request:44129", "accept": false, - "reason": "GPT-J/CodeGen and MPT are different concrete changes." + "reason": "DeBERTa and SpeechT5 are different model families with separate refactors." }, { - "left": "pull_request:44018", - "right": "pull_request:44066", + "left": "pull_request:44044", + "right": "pull_request:44154", "accept": false, - "reason": "GPT-Neo is separate from GPT-J/CodeGen, despite a similar output-tracing refactor pattern." + "reason": "DeBERTa and VITS are unrelated model-specific output-tracing changes." }, { - "left": "pull_request:44019", + "left": "pull_request:44066", "right": "pull_request:44722", - "accept": false, - "reason": "ResNet and GPT-J/CodeGen are unrelated model changes." + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor in the same code path; 44722 looks like the broader overlapping iteration." }, { - "left": "pull_request:44068", + "left": "pull_request:44085", "right": "pull_request:44722", - "accept": false, - "reason": "GPT-Neo and GPT-J/CodeGen are different underlying fixes." - }, - { - "left": "pull_request:44098", - "right": "pull_request:44101", - "accept": false, - "reason": "ViLT and Flaubert/XLM are different models and code paths." - }, - { - "left": "pull_request:44073", - "right": "pull_request:44101", - "accept": false, - "reason": "VisualBERT and Flaubert/XLM are unrelated model-specific refactors." - }, - { - "left": "pull_request:44056", - "right": "pull_request:44161", - "accept": false, - "reason": "MPNet and LongT5 are different model families." - }, - { - "left": "pull_request:44068", - "right": "pull_request:44086", - "accept": false, - "reason": "GPT-Neo and MGP-STR are separate model-specific changes." + "accept": true, + "reason": "Same GPT-J output-tracing change in the same file; 44722 incorporates that change and extends it to CodeGen." }, { - "left": "pull_request:44098", - "right": "pull_request:44161", + "left": "pull_request:44085", + "right": "pull_request:44129", "accept": false, - "reason": "ViLT and LongT5 are unrelated model-specific changes." + "reason": "GPT-J and SpeechT5 are unrelated code paths." }, { - "left": "pull_request:44073", - "right": "pull_request:44161", + "left": "pull_request:43996", + "right": "pull_request:44007", "accept": false, - "reason": "VisualBERT and LongT5 are different model families." + "reason": "Different model families and different output-handling implementations." }, { - "left": "pull_request:44059", - "right": "pull_request:44086", + "left": "pull_request:43996", + "right": "pull_request:44129", "accept": false, - "reason": "GPT-2 and MGP-STR are unrelated model-specific changes." + "reason": "CVT/FNet and SpeechT5 are unrelated model implementations." }, { - "left": "pull_request:44085", - "right": "pull_request:44086", + "left": "pull_request:44044", + "right": "pull_request:44129", "accept": false, - "reason": "GPT-J and MGP-STR are unrelated model fixes." + "reason": "DeBERTa and SpeechT5 are distinct model families." }, { - "left": "pull_request:44098", - "right": "pull_request:44101", + "left": "pull_request:44044", + "right": "pull_request:44154", "accept": false, - "reason": "ViLT and Flaubert/XLM are different model implementations." + "reason": "DeBERTa and VITS are different code paths; same refactor pattern only." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is grounded in the packet and stays conservative: most PRs are indeed model-specific output-tracing refactors tied thematically to issue #43979, while the accepted soft edges are limited to the GPT-Neo/GPT-J subgroup where the diffs and filenames make same-bugfix duplication plausible. The representative choice of PR 44722 is also supported as the later, broader GPT-J/CodeGen version." + "accept": false, + "feedback": "The cluster theme is mostly grounded, but the summary overstates the evidence by saying all PRs are \"orbiting\" #43979; several items in the packet have no explicit issue link, so that should be softened to \"many/most\" or \"the relevant PRs.\" The soft-edge verdicts are generally conservative, and the same-model duplicate calls look plausible, but I would still tighten the umbrella claim before accepting." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -19338,7 +19123,7 @@ { "label": "pull_request", "component_index": 14, - "component_count": 28, + "component_count": 27, "review_unit_index": 2, "review_unit_count": 2, "status": "reviewed", @@ -19365,14 +19150,14 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "e0ccc20ec952865f221a251a79dbb7fabb9bcf2bf79f796ac2a19334807f878b", + "prepared_review_unit_hash": "f38a7ba969b9d7759b3af9963b1a1d2f9ab5d677b8533c1d9bdac70f3c509405", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 78, - "serialized_chars": 84894, - "estimated_input_tokens": 21224, - "estimated_eval_tokens": 42704 + "serialized_chars": 84899, + "estimated_input_tokens": 21225, + "estimated_eval_tokens": 42706 }, "budget": { "node_count": 8, @@ -19387,48 +19172,48 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These PRs are all part of the same broad \u201cstandardized output tracing\u201d migration pattern, but they are not duplicates: each one targets a different model and code path. None of the soft edges should be merged as the same underlying change.", - "confidence": 0.94, + "summary": "These PRs are all related to a common standardized output-tracing refactor pattern, but they touch different model implementations and distinct code paths, so they are not duplicates of one another. I would treat them as separate PRs in the same thematic cluster.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "Pick PR 44129 as the canonical PR: it is the most clearly scoped and the only one with an explicit issue target, making it the best representative of this output-tracing refactor series.", + "canonical_pr_reason": "44129 is the strongest representative because it has the clearest scoped refactor, an explicit linked issue target, and concrete output-tracing changes in one model. That said, it is still not a duplicate of the others.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the strongest standalone candidate because it has a tracked issue link and a concrete SpeechT5-specific tracing change rather than just a generic migration pattern.", + "best_pr_reason": "44129 is a good cluster representative since it is a focused model-specific implementation of the shared output-capture refactor theme and includes an explicit issue reference.", "soft_edge_verdicts": [ { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "Same refactor theme, but FocalNet vs ImageGPT are different model implementations and different forward paths." + "reason": "Both are standardized-output refactors, but one is for FocalNet and the other for ImageGPT; they change different model code paths and are not mergeable as one PR without becoming an unrelated multi-model sweep." }, { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "Both are output-capture refactors, but VisualBERT and TextNet are separate model code paths and not the same bug/change." + "reason": "VisualBERT and TextNet are separate implementations. The similarity is the refactor pattern, not the underlying bug/change in a single code path." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "MPNet and VisualBERT each need their own tracing migration; they do not look mergeable into one concrete fix." + "reason": "These target different models (MPNet vs VisualBERT) and different tracing hooks, so they are not the same concrete fix." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "Shared pattern only; MPNet and TextNet are distinct models with separate output semantics." + "reason": "MPNet and TextNet are unrelated model files with separate output-tracing refactors; same theme, different changes." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBERT are unrelated model implementations, so this is not the same underlying change." + "reason": "RegNet and SqueezeBERT are distinct model-specific migrations to the shared output-capture pattern, not the same underlying bug or change." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "Both refactor output handling, but SqueezeBERT and SpeechT5 fix different model-specific code paths and should stay separate." + "reason": "SqueezeBERT and SpeechT5 modify different model forward paths. Sharing the standardized output-capture goal is not enough to treat them as duplicates." } ] }, @@ -19442,7 +19227,7 @@ { "label": "pull_request", "component_index": 15, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19514,298 +19299,298 @@ "pull_request:44002|pull_request:46106", "pull_request:43998|pull_request:46106" ], - "prepared_review_unit_hash": "a5e57acf87db12fb63ff9d6c0bdbc2b6234bc5a6aeed866109b287236bbf60e5", + "prepared_review_unit_hash": "bfbd37308589ee199874faa3eaa56ea422b593f808c719179147d4323e46ccd4", "component_budget": { "node_count": 14, "item_count": 14, "soft_pair_count": 45, - "serialized_chars": 43785, - "estimated_input_tokens": 10947, - "estimated_eval_tokens": 22150 + "serialized_chars": 43783, + "estimated_input_tokens": 10946, + "estimated_eval_tokens": 22148 }, "budget": { "node_count": 14, "item_count": 14, "soft_pair_count": 45, - "serialized_chars": 43785, - "estimated_input_tokens": 10947, - "estimated_eval_tokens": 22150 + "serialized_chars": 43783, + "estimated_input_tokens": 10946, + "estimated_eval_tokens": 22148 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are mostly independent, model-specific output-tracing refactors plus one unrelated docs-only toctree fix. They share a theme, but they do not look like the same underlying bug/change, so the soft duplicate edges should be rejected.", - "confidence": 0.95, + "summary": "This cluster is a set of separate model-specific PRs that all follow the same output-tracing refactor pattern, but they do not fix the same concrete code path. I reject the soft duplicate edges and treat them as distinct PRs. PR 44025 is the best representative of the cluster.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44003 is the broadest and most representative refactor in this set: it spans multiple model files and introduces shared tracing plumbing, so it is the best anchor if one PR must represent the cluster.", + "canonical_pr_reason": "PR 44025 is the clearest representative: it is a self-contained output-tracing refactor, spans two closely related files, and directly reflects the cluster\u2019s common migration pattern.", "best_issue_reason": null, - "best_pr_reason": "PR 44003 is the strongest representative of the shared theme because it makes the most substantial tracing-related code change; the others are narrower per-model refactors, and 46106 is just a docs toctree edit.", + "best_pr_reason": "PR 44025 is the best single PR to anchor the cluster because it is explicit, focused, and representative without being overly specialized to one unusual code path.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different models and files (`timm_backbone` vs `upernet`); same refactor theme, but not the same concrete bug/change." + "reason": "Different model implementations and forward paths; same refactor theme, but not the same mergeable change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "`timm_backbone` and `vision_text_dual_encoder` are unrelated code paths; these are separate output-tracing refactors." + "reason": "Different architectures (`timm_backbone` vs `vision_text_dual_encoder`); similar output-tracing cleanup only." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "`timm_backbone` and `univnet` touch different architectures and forward signatures; not one duplicate change." + "reason": "Separate model code paths; not the same underlying bug or fix." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Different model families and files; 44003 adds tracing plumbing for Mamba, not the same change as timm_backbone." - }, - { - "left": "pull_request:43998", - "right": "pull_request:44004", - "accept": false, - "reason": "`timm_backbone` vs `codegen` are different subsystems and code paths." + "reason": "Both are output-tracing refactors, but they target different model families and concrete implementations." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Different models (`timm_backbone` vs `rwkv`); shared theme is too broad to treat as duplicate." - }, - { - "left": "pull_request:43998", - "right": "pull_request:46106", - "accept": false, - "reason": "A model refactor and a docs toctree change are clearly unrelated." + "reason": "Unrelated model-specific forward paths; too broad to be one duplicate PR." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "Different model implementations (`mobilenet_v1` vs `univnet`); same tracing motif, not the same bug or patch." + "reason": "Different model code paths (`mobilenet_v1` vs `univnet`); only the refactor style overlaps." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "Different files and model code paths; both refactor output handling but independently." + "reason": "Different model families and outputs; not the same concrete change." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "`mobilenet_v1` and `depth_anything` are separate model refactors; not mergeable as one concrete fix." + "reason": "Separate models with distinct forward logic; same tracing migration, not one PR-worthy fix." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "Different encoder-decoder modules and different implementations; only the general topic overlaps." + "reason": "Different model code paths; the similarity is only the output-tracing refactor pattern." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "`mobilenet_v1` and `speech_encoder_decoder` are unrelated model paths; not a duplicate." + "reason": "No shared underlying bug or code path beyond the general refactor theme." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "`vision_text_dual_encoder` and `univnet` are different models and code paths; same refactor pattern only." + "reason": "Different architectures (`vision_text_dual_encoder` vs `univnet`); not mergeable as one fix." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different model families (`vision_text_dual_encoder` vs `upernet`); no shared concrete bug." + "reason": "Different model implementations; only a shared migration approach." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "Separate model refactors; 44003 changes Mamba-specific tracing, not the vision-text dual encoder." - }, - { - "left": "pull_request:44000", - "right": "pull_request:44004", - "accept": false, - "reason": "`vision_text_dual_encoder` and `codegen` are unrelated implementations despite similar decorator changes." + "reason": "Distinct model families and forward methods; not the same concrete problem." }, { "left": "pull_request:44000", "right": "pull_request:44025", "accept": false, - "reason": "Different model code and different tracing adjustments; not one underlying issue." + "reason": "Both touch output handling, but on unrelated model code paths." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "`univnet` and `upernet` are separate model code paths with no evidence of a shared bug." + "reason": "Separate models and outputs; same refactor pattern is not enough to duplicate them." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Different architecture families (`univnet` vs `mamba`); both refactor output tracing but are not the same change." + "reason": "Different model stacks (`univnet` vs `falcon_mamba/mamba`); not one underlying fix." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "`univnet` and `depth_anything` are distinct models; not plausible as one merged PR." + "reason": "Different code paths and different model families; only broadly similar tracing cleanup." }, { "left": "pull_request:44002", - "right": "pull_request:44003", + "right": "pull_request:44029", "accept": false, - "reason": "`upernet` and `mamba` touch unrelated code paths; the overlap is only the tracing refactor style." + "reason": "Different model implementations; not the same bug or change." }, { "left": "pull_request:44002", - "right": "pull_request:44004", + "right": "pull_request:44003", "accept": false, - "reason": "Different models and filenames; no shared concrete fix." + "reason": "Separate model code paths; same general refactor class only." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "`mamba`/`falcon_mamba` tracing changes are unrelated to `superpoint` feature extraction code." + "reason": "Different model families (`mamba` vs `superpoint`) and distinct output structures." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "`mamba` and `codegen` are different model implementations with different call signatures and tracing behavior." + "reason": "Codegen refactor is a different concrete change than the Mamba/Falcon-Mamba tracing work." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model families and forward paths; similar refactor intent only." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "Different code paths (`univnet` vs `codegen`); not a duplicate change." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model-specific output handling; no single shared bug to merge." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "`depth_anything` and `vision_encoder_decoder` are different model families; same output-tracing theme only." + "reason": "Different models (`depth_anything` vs `vision_encoder_decoder`) and different forward logic." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "`depth_anything` and `speech_encoder_decoder` are separate refactors, not the same code-path problem." + "reason": "Different model wrappers and output signatures; same refactor pattern only." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "`depth_anything` and `superpoint` are unrelated model implementations." + "reason": "Unrelated model code paths; not the same underlying change." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "Different model families (`depth_anything` vs `rwkv`); no duplicate evidence beyond general tracing work." + "reason": "Different architectures and output plumbing; too broad to count as one duplicate." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "`depth_anything` and `dpr` are distinct models and code paths; not the same underlying fix." + "reason": "Different model code paths; only the output-tracing migration overlaps." }, { "left": "pull_request:44026", - "right": "pull_request:44027", + "right": "pull_request:44029", "accept": false, - "reason": "`vision_encoder_decoder` and `speech_encoder_decoder` are separate modules with analogous but independent refactors." + "reason": "Different model families and forward implementations; not one fix." }, { "left": "pull_request:44026", - "right": "pull_request:44028", + "right": "pull_request:44027", "accept": false, - "reason": "Different model code; no sign they fix the same bug." + "reason": "Different encoder-decoder wrappers; same broad refactor theme, but not the same code path." }, { "left": "pull_request:44026", - "right": "pull_request:44029", + "right": "pull_request:44028", "accept": false, - "reason": "`vision_encoder_decoder` and `rwkv` are unrelated architectures and forward paths." + "reason": "Different model implementations and output types; not mergeable into one PR." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "Different models (`vision_encoder_decoder` vs `dpr`); overlapping theme is too broad." + "reason": "Distinct model code paths; shared tracing refactor only." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "`speech_encoder_decoder` and `superpoint` are unrelated model code paths." + "reason": "Speech encoder-decoder and SuperPoint are unrelated concrete changes." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Different architectures (`speech_encoder_decoder` vs `rwkv`); not a duplicate." + "reason": "Different model families and output handling; not a duplicate." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "`speech_encoder_decoder` and `dpr` touch separate models and issues." + "reason": "Separate model-specific forward paths; only the refactor style is similar." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "`superpoint` and `rwkv` are unrelated model implementations." + "reason": "Different models and outputs; no shared underlying bug." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "Different files and model behavior; no shared concrete bug/change." + "reason": "Unrelated model code paths; same broad output-tracing refactor only." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "`rwkv` and `dpr` are different model families; same output-tracing motif only." + "reason": "Different model families and internal logic; not the same concrete fix." }, { "left": "pull_request:44027", "right": "pull_request:46106", "accept": false, - "reason": "A speech-encoder-decoder refactor and a docs toctree edit are unrelated." + "reason": "Documentation/toctree change is unrelated to the speech encoder-decoder tracing refactor." }, { "left": "pull_request:44026", "right": "pull_request:46106", "accept": false, - "reason": "`vision_encoder_decoder` code refactor vs docs navigation change; not a duplicate." + "reason": "Docs toctree cleanup is not the same as the vision-encoder-decoder code change." }, { "left": "pull_request:44001", "right": "pull_request:46106", "accept": false, - "reason": "Model tracing refactor and documentation-only toctree change are clearly unrelated." + "reason": "Unrelated docs navigation edit versus a model tracing refactor." }, { "left": "pull_request:44002", "right": "pull_request:46106", "accept": false, - "reason": "`upernet` code changes do not match a docs-only fix." + "reason": "Documentation tree update is unrelated to UperNet output tracing." }, { "left": "pull_request:43998", "right": "pull_request:46106", "accept": false, - "reason": "`timm_backbone` refactor and docs toctree cleanup are unrelated." + "reason": "Docs-only toctree change is unrelated to the timm_backbone refactor." } ] }, @@ -19819,7 +19604,7 @@ { "label": "pull_request", "component_index": 16, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19838,52 +19623,52 @@ "pull_request:44662|pull_request:45075", "pull_request:45075|pull_request:45401" ], - "prepared_review_unit_hash": "d988754a07fa4029d35b7c1e83638e614961d6eef9d8aecb45d551c1a06b8250", + "prepared_review_unit_hash": "97016c49c9930f8896c30fdcf6ff36a2f98b26e014812c5b3364d146e0126788", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10915, - "estimated_input_tokens": 2729, - "estimated_eval_tokens": 5714 + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 }, "budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10915, - "estimated_input_tokens": 2729, - "estimated_eval_tokens": 5714 + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-integration PRs, not duplicates: PenguinVL, DeepSeek-OCR-2, and Voxtral TTS. They overlap mainly on the standard Transformers boilerplate for docs and auto mappings, but they target different models and codepaths.", - "confidence": 0.97, + "summary": "All three PRs are standalone model-addition changes for different models (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They share only framework boilerplate and auto-registration patterns, not the same underlying feature or code-path fix.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: each PR adds a different model implementation, so they are not the same underlying change.", + "canonical_pr_reason": "No true canonical PR: these are separate model implementation PRs for different architectures, so there is no duplicate to collapse into one.", "best_issue_reason": null, - "best_pr_reason": "No single best PR for a duplicate cluster. If forced to pick a representative, 45075 is the most complete and mature, but it is still a distinct model addition rather than a duplicate of the others.", + "best_pr_reason": "PR 45075 is the strongest standalone representative: it has the broadest implementation surface (core model, processing, auto mappings, tests, docs, and conversion helpers) and a clear, self-contained scope.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different model additions (PenguinVL vs Voxtral TTS); shared files are only registration/docs boilerplate, not the same code-path fix." + "reason": "Different models and different functionality (PenguinVL vs Voxtral TTS). Shared files are only registration/docs boilerplate, not the same bug or feature." }, { "left": "pull_request:44662", "right": "pull_request:45075", "accept": false, - "reason": "Different model additions (PenguinVL vs DeepSeek-OCR-2); overlap is limited to common Transformers scaffolding." + "reason": "Both add model support, but for unrelated models with different code paths and artifacts; this is broad subsystem overlap, not a duplicate change." }, { "left": "pull_request:45075", "right": "pull_request:45401", "accept": false, - "reason": "Different model additions (DeepSeek-OCR-2 vs Voxtral TTS); they do not appear to fix or implement the same concrete change." + "reason": "DeepSeek-OCR-2 and Voxtral TTS are distinct model integrations with different processing and modeling logic; they are not plausibly mergeable into one PR as the same change." } ] }, @@ -19897,7 +19682,7 @@ { "label": "pull_request", "component_index": 17, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19913,12 +19698,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "8f0594ef554ba3a9cd417ac6d15c34cd4015116a67c74bb1a56c719cffef59b2", + "prepared_review_unit_hash": "e6d9a11e0ed8bd4021ac41e680dc1615b7bf9fedf589a013bdd75ea25af1680c", "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 }, @@ -19926,7 +19711,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4407, + "serialized_chars": 4408, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -19935,18 +19720,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related by the broad theme of fusion, but they target different layers: one adds a generic module-fusion API, while the other wires up n-to-1 kernel fusion via KernelConfig and integration changes. I would not treat them as duplicates.", - "confidence": 0.84, + "summary": "These PRs are related in the broad area of module/kernel fusion, but they are not the same change: one adds a standalone Module Fusion API scaffold, while the other wires up n-to-1 kernel fusion through KernelConfig and integration points. I would not treat them as duplicates.", + "confidence": 0.86, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45363 is the better cluster anchor because it is the more concrete end-to-end feature implementation, spanning KernelConfig, model loading, and integration hooks for n-to-1 kernel fusion.", + "canonical_pr_reason": "PR #45363 is the better canonical PR because it is the more complete, user-facing feature implementation with multiple integration changes, whereas #44979 is mainly the lower-level Module Fusion API foundation.", "best_issue_reason": null, - "best_pr_reason": "PR 45363 is the best representative of a user-visible change because it defines the actual kernel-fusion workflow and includes usage/example integration, whereas PR 44979 is a lower-level API addition for a different fusion abstraction.", + "best_pr_reason": "PR #45363 is the best representative of the cluster\u2019s overall theme since it delivers the broader kernel-fusion workflow; #44979 is supporting infrastructure rather than the same concrete change.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "Related only at a high level. PR 44979 adds a general module-fusion API and new module_fusion.py infrastructure, while PR 45363 implements n-to-1 kernel fusion through KernelConfig plus loader/integration changes. They do not look like the same underlying bug or a single PR-sized change." + "reason": "Related fusion infrastructure, but not the same concrete bug/change and not something that should be merged into a single PR." } ] }, @@ -19960,7 +19745,7 @@ { "label": "pull_request", "component_index": 18, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -19976,7 +19761,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "f3bb3c28164764831b9e37df3862b5784dff9a36d1a3255950b2db9f1ef340ec", + "prepared_review_unit_hash": "724597b27a3cef28d0c2128dc1553b048b6911a327270a16447c8982e6183007", "component_budget": { "node_count": 2, "item_count": 2, @@ -19998,18 +19783,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests address the same Mistral regex patch AttributeError in the same file and target the same issue; one is a narrower code-only fix and the other adds the same fix plus tests, so they look like duplicate PRs for the same bug.", - "confidence": 0.96, - "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same fix as 45086 plus test coverage and a clearer title/body, making it the more complete implementation of the shared bugfix.", - "best_issue_reason": null, - "best_pr_reason": "PR 45317 is the strongest representative of the cluster: it fixes the concrete AttributeError in _patch_mistral_regex and also includes regression tests, so it is the most complete and mergeable version.", + "summary": "Both pull requests address the same Mistral regex patching AttributeError in the same code path and target the same issue; 45317 is the more complete variant because it adds tests.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #45081 is the shared explicit target for both PRs, so it is the natural canonical issue.", + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same fix plus test coverage, making it the more complete and likely final form of the change.", + "best_issue_reason": "#45081 is the only explicit issue target and represents the underlying bug both PRs are trying to resolve.", + "best_pr_reason": "pull_request:45317 is the best representative PR: same fix, same file, and it adds tests to validate the AttributeError regression.", "soft_edge_verdicts": [ { "left": "pull_request:45086", "right": "pull_request:45317", "accept": true, - "reason": "Same underlying code-path bug in src/transformers/tokenization_utils_tokenizers.py, same explicit issue target, and the diff shows the same AttributeError fix; 45317 appears to be a superset of 45086 with added tests, so these are duplicate PRs for one change." + "reason": "Same underlying AttributeError fix in _patch_mistral_regex, same target issue #45081, and the diffs overlap on the exact code-path; 45317 is a stricter superset with tests, so these are duplicate/same-change PRs." } ] }, @@ -20023,7 +19808,7 @@ { "label": "pull_request", "component_index": 19, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20039,7 +19824,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "f8c8e10594b2058eeb8c217209c1f402d26b90b7bc580b77cc21a9efd0e1677c", + "prepared_review_unit_hash": "1b71047d5bf36b9aca79b73d2fb8e0308e74f8b85cecaad6793215abd9f32f6d", "component_budget": { "node_count": 2, "item_count": 2, @@ -20061,18 +19846,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related NemotronH architecture PRs, but they are not duplicates. PR #45569 is a broader, more complete change introducing NemotronH dense/sparse variants plus docs and auto-mapping updates, while PR #45115 is a narrower refactor around inheriting from GraniteMoeHybrid and key conversion mapping.", - "confidence": 0.86, + "summary": "These PRs are related to NemotronH internals, but they address different changes: one is a refactor/inheritance and checkpoint conversion mapping update, while the other introduces the dense/sparse model split with docs, auto-mapping, and tests.", + "confidence": 0.89, "canonical_issue_reason": null, - "canonical_pr_reason": "#45569 is the broader and more complete NemotronH change: it adds the dense/sparse model split, updates docs and auto mappings, and includes test/model code across the new variants.", + "canonical_pr_reason": "PR 45569 is the better canonical representative because it is the broader, more complete feature change: it adds the new NemotronH dense/sparse model structure, updates docs and auto-mappings, and includes tests. PR 45115 is a narrower refactor focused on GraniteMoeHybrid-style inheritance and conversion renames.", "best_issue_reason": null, - "best_pr_reason": "#45569 is the best representative of the cluster because it captures the full architectural rework, not just the inheritance/refactor portion.", + "best_pr_reason": "PR 45569 is the strongest standalone PR in the cluster since it carries the main user-visible/model-structure work and supporting documentation/tests, whereas PR 45115 is a narrower implementation refactor.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "They overlap on NemotronH internals, but the concrete changes differ: #45115 focuses on conversion/inheritance refactoring, while #45569 implements the larger dense/sparse split and documentation/model wiring. They are related, not the same change." + "reason": "Reject: despite overlapping NemotronH files, the PRs are not the same underlying change. 45115 is about refactoring NemotronH to inherit GraniteMoeHybrid and adding conversion-key renames, while 45569 is about introducing NemotronH dense/sparse variants plus docs, auto-mapping, and tests. They could coexist, but they are not one mergeable duplicate PR." } ] }, @@ -20086,7 +19871,7 @@ { "label": "pull_request", "component_index": 20, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20102,7 +19887,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "970187413786ab830c06a918addcd24318e1546fea5d6c4e5e67b6b526e8bb16", + "prepared_review_unit_hash": "2c1454679a95544862b9541330e98e2cec70c98157195ee5865798e5ca03dcc6", "component_budget": { "node_count": 2, "item_count": 2, @@ -20124,18 +19909,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related feature PRs in the new heterogeneity area, but they are not duplicates: one adds per-layer config support and the other adds broader heterogeneous modeling/cache integration on top of that foundation.", - "confidence": 0.88, + "summary": "Two related but distinct feature PRs: one adds heterogeneous config support, the other adds heterogeneous model/runtime support. They overlap in shared config helpers, but they do not look like the same mergeable change.", + "confidence": 0.79, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #45332 is the broader umbrella change for heterogeneous model support and includes the modeling-side integration, so it is the best representative of the cluster, but it is not the same change as #45333.", + "canonical_pr_reason": "PR #45332 is the more canonical umbrella change: it implements the concrete model/runtime behavior for heterogeneous layers and includes the cache/modeling updates that depend on config support.", "best_issue_reason": null, - "best_pr_reason": "PR #45332 is the strongest single representative because it spans the end-to-end heterogeneous model support, while #45333 is a narrower config-layer prerequisite.", + "best_pr_reason": "PR #45332 is the strongest standalone candidate because it covers the main end-user behavior change, not just the supporting config plumbing.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "They share the heterogeneity feature area and some config files, but #45333 is specifically per-layer config support while #45332 adds broader modeling/cache behavior; they are related components, not the same concrete fix/change." + "reason": "These are related layers of the same feature area, but one is config plumbing and the other is model/runtime support; they do not look like the same concrete code-path problem that could be merged into a single PR." } ] }, @@ -20149,7 +19934,7 @@ { "label": "pull_request", "component_index": 21, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20165,12 +19950,12 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "b653c1debfed4c5a5df0ace218957f7d146a0cfdd0ae67886a404bcc6147ae78", + "prepared_review_unit_hash": "88128e6ef8b67744d07c06e7f912a37f00472f2d4c421b82781fa87f5f88c598", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5513, + "serialized_chars": 5514, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -20178,7 +19963,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5513, + "serialized_chars": 5514, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -20187,18 +19972,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are both typing-related PRs in transformers, but they are not the same change: one is a broad type-checking sweep with pyproject ignore-policy updates, while the other adds concrete typing protocols and narrow code support for specific modules. They should not be deduplicated.", - "confidence": 0.83, + "summary": "Both PRs are typing-focused cleanups, but they are not the same concrete change: #45415 is a broad repo-wide type-checking pass with pyproject typing-ignore adjustments, while #45425 adds specific typing helper protocols/aliases and related modeling-utils annotations. They share some files, but the underlying work is different enough that they should not be merged as duplicates.", + "confidence": 0.79, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45425 is the more concrete, self-contained typing refactor; PR 45415 is a much broader repository-wide type-checking sweep and ignore-config change, so it is a poorer representative of the cluster.", + "canonical_pr_reason": "#45415 is the broader umbrella typing cleanup and best represents the cluster\u2019s overall theme; #45425 is a narrower, more specialized follow-up-style typing change.", "best_issue_reason": null, - "best_pr_reason": "PR 45425 is the best representative for a typing-fix cluster because it introduces explicit protocol/type definitions and targeted code changes, whereas PR 45415 is mostly a large mechanical type-checking pass across many files.", + "best_pr_reason": "#45415 is the best representative PR because it covers the widest scope of the typing initiative and includes the typing-policy change in pyproject.toml, making it the most canonical cluster member.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Reject: both are typing-focused, but they address different underlying changes. 45415 is a broad `src/transformers/*py` type-checking sweep plus pyproject ignore adjustments; 45425 adds specific typing protocols and related support in a narrower set of files. They are not the same concrete bug/change and would not plausibly merge as one PR." + "reason": "Shared typing-related files are not enough: #45415 is a large multi-file type-checking sweep, while #45425 introduces new typing protocols/aliases for specific internals. They do not fix the same concrete bug or merge cleanly into one PR." } ] }, @@ -20212,86 +19997,7 @@ { "label": "pull_request", "component_index": 22, - "component_count": 28, - "review_unit_index": 1, - "review_unit_count": 1, - "status": "reviewed", - "reason": null, - "source": "llm", - "cache_hit": false, - "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-45487-4", - "nodes": [ - "pull_request:45487", - "pull_request:46109", - "pull_request:46135", - "pull_request:46151" - ], - "soft_pairs": [ - "pull_request:45487|pull_request:46151", - "pull_request:46109|pull_request:46151", - "pull_request:46135|pull_request:46151" - ], - "prepared_review_unit_hash": "c0153906757c081b3467067dfe10746dd71b0cd9b9935961a99b07b0887a7937", - "component_budget": { - "node_count": 4, - "item_count": 4, - "soft_pair_count": 3, - "serialized_chars": 9913, - "estimated_input_tokens": 2479, - "estimated_eval_tokens": 5214 - }, - "budget": { - "node_count": 4, - "item_count": 4, - "soft_pair_count": 3, - "serialized_chars": 9913, - "estimated_input_tokens": 2479, - "estimated_eval_tokens": 5214 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": false, - "analyst_result": { - "summary": "This cluster is heterogeneous: one substantive model-code bug fix (AltCLIP/ChineseCLIP device placement) and three unrelated test-only PRs in different areas. I would not merge any soft edges as duplicates.", - "confidence": 0.96, - "canonical_issue_reason": null, - "canonical_pr_reason": "pull_request:45487 is the only PR that fixes a concrete runtime code-path bug; the others are separate test expectation changes in unrelated subsystems.", - "best_issue_reason": null, - "best_pr_reason": "pull_request:45487 is the best representative because it addresses an actual model-parallel correctness issue rather than a test-only update.", - "soft_edge_verdicts": [ - { - "left": "pull_request:45487", - "right": "pull_request:46151", - "accept": false, - "reason": "45487 fixes AltCLIP/ChineseCLIP token-type device placement in model code; 46151 only adjusts token-classification pipeline test expectations. Different subsystems and different change types." - }, - { - "left": "pull_request:46109", - "right": "pull_request:46151", - "accept": false, - "reason": "Both are test maintenance, but one is distributed trainer parity tests and the other is pipeline token-classification expectations. They do not target the same underlying bug or change." - }, - { - "left": "pull_request:46135", - "right": "pull_request:46151", - "accept": false, - "reason": "46135 adds XPU expectations for Kosmos2.5 integration tests, while 46151 updates token-classification pipeline expectations. Same broad theme of test updates, but not the same code-path problem." - } - ] - }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, - "error_kind": null, - "error_message": null - }, - { - "label": "pull_request", - "component_index": 23, - "component_count": 28, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20309,12 +20015,12 @@ "pull_request:45513|pull_request:46084", "pull_request:46084|pull_request:46098" ], - "prepared_review_unit_hash": "324a21a4843bb9bf1771b14cb1684bdd88e514e006788fe7a8c78829d520c5ea", + "prepared_review_unit_hash": "ce1bf896e68f5dd02102fd168710da03ac3ce91c513e0de8852e7c74b5c8f9d1", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 7428, + "serialized_chars": 7427, "estimated_input_tokens": 1857, "estimated_eval_tokens": 3970 }, @@ -20322,7 +20028,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 7428, + "serialized_chars": 7427, "estimated_input_tokens": 1857, "estimated_eval_tokens": 3970 }, @@ -20331,30 +20037,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "PR 46084 and PR 46098 are both Mamba2 cache/seq_len>1 fixes for the same underlying bug, with 46098 appearing to be the broader superset. PR 45513 is a separate Qwen/Olmo cached-forward fix and should not be linked to the Mamba2 work.", - "confidence": 0.87, + "summary": "Two Mamba2 PRs appear to address the same use_cache seq_len>1 bug; the Qwen3.5 PR is unrelated and should stay separate. The broader Mamba2 PR is the best representative of that fix.", + "confidence": 0.78, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46098 is the strongest canonical PR: it targets the exact Mamba2 seq_len>1 cache bug, explicitly links the tracking issue, and covers the broader CPU/GPU fix with tests and docs.", + "canonical_pr_reason": "PR #46098 is the best canonical representative because it directly targets the Mamba2 use_cache seq_len>1 incorrect-results bug, covers both CPU/GPU paths, and includes tests/docs, making it the most complete version of the fix.", "best_issue_reason": null, - "best_pr_reason": "PR 46098 is the best standalone representative because it is the most complete and explicit fix for the underlying Mamba2 use_cache > 1-token problem.", + "best_pr_reason": "PR #46098 is the strongest cluster representative: it is explicit about the underlying bug, has the same Mamba2 code path, and is broader/more complete than the narrower GPU-focused predecessor PR #46084.", "soft_edge_verdicts": [ { "left": "pull_request:45513", "right": "pull_request:46084", "accept": false, - "reason": "Different models and different bugs: Qwen3.5/Olmo GDN linear attention cached forward versus Mamba2 cache handling. Similar wording about cached multi-token behavior is not enough to treat these as the same fix." + "reason": "Different model families and different implementations; they only share a generic cached-forward symptom, not the same concrete bug or mergeable change." }, { "left": "pull_request:46084", "right": "pull_request:46098", "accept": true, - "reason": "Both address the same Mamba2 use_cache with seq_len > 1 failure mode against the same code path family and issue. 46098 looks like a broader version of the same fix, so they could plausibly be merged into one PR." + "reason": "Both target the same Mamba2 use_cache seq_len>1 bug in modeling_mamba2.py; #46098 looks like a broader follow-up/superset with the same underlying fix and could plausibly be merged into one PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: PR 45513 is clearly a separate Qwen3.5/Olmo hybrid cached-forward change, while PRs 46084 and 46098 both target the same Mamba2 use_cache / seq_len > 1 bug and share the same issue target (46032). The verdict that 46084 and 46098 are mergeable into one Mamba2 bugfix PR is conservative enough given the overlapping files and issue linkage. No evidence of overstatement." + "feedback": "The summary is grounded in the packet: PR #45513 is a different model family and does not look mergeable with the Mamba2 fix, while PRs #46084 and #46098 share the same issue target and Mamba2 files, so treating #46098 as the broader canonical representative is conservative. No evidence is overstated." }, "evaluator_used": true, "retried": false, @@ -20364,8 +20070,8 @@ }, { "label": "pull_request", - "component_index": 24, - "component_count": 28, + "component_index": 23, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20381,12 +20087,12 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "59395b4517f1340659cd594a78987b9e96983d6173b9c2bb97a4bc9a87904041", + "prepared_review_unit_hash": "738cdf6fadb059a2922d512c4a817a4753d3e7ad472b9fad9596fb34b13a705e", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4507, + "serialized_chars": 4508, "estimated_input_tokens": 1127, "estimated_eval_tokens": 2510 }, @@ -20394,7 +20100,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4507, + "serialized_chars": 4508, "estimated_input_tokens": 1127, "estimated_eval_tokens": 2510 }, @@ -20403,18 +20109,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related EP-loading PRs, but they fix different bugs on different code paths: one targets config serialization and EP detection for accelerate/ZeRO-3, the other fixes parameter wrapping to stop expert weights being overwritten under FSDP2.", - "confidence": 0.93, + "summary": "Both PRs touch expert-parallel loading, but they fix different bugs: one addresses missing distributed config/EP detection for ZeRO-3 + accelerate launch, while the other fixes experts being overwritten during FSDP2 sharding/broadcast. They are related but not duplicates.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45662 is the more concrete, narrowly-scoped bug fix: it addresses a specific expert-parameter overwrite during loading and has a clear runtime failure mode. PR 45548 is related to EP setup, but it solves a different problem around config persistence and EP detection.", + "canonical_pr_reason": "PR #45662 is the better canonical representative: it targets a concrete, specific code-path failure (expert weights being overwritten during distributed loading) and includes the broader set of loading-path changes needed for that bug.", "best_issue_reason": null, - "best_pr_reason": "PR 45662 is the strongest representative of the cluster because it fixes a well-defined data-corruption/loading issue in a specific code path, whereas PR 45548 is more about enabling EP-aware loading behavior.", + "best_pr_reason": "PR #45662 is the strongest standalone PR in the cluster because it fixes a clearly defined runtime bug with broader implementation coverage; #45548 is narrower and more about enabling EP metadata/detection for a different loading path.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "Both concern expert parallelism, but they do not look like the same underlying bug. 45548 changes config serialization and adds an EP-detection property for accelerate/ZeRO-3 loading; 45662 changes loading/wrapping logic to prevent experts from being overwritten under FSDP2. Different failure modes, different fixes, not mergeable as one PR." + "reason": "Different underlying bugs and fix paths: #45548 is about EP + DeepSpeed ZeRO-3 loading via accelerate launch and serialization of distributed_config/has_ep, while #45662 fixes EP + FSDP2 experts being silently overwritten by rank-0 broadcast via post-shard wrapping. Shared subsystem only, not the same change." } ] }, @@ -20427,8 +20133,8 @@ }, { "label": "pull_request", - "component_index": 25, - "component_count": 28, + "component_index": 24, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20444,40 +20150,40 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "e284af41cce128046c2953c07701dc08f0dabf72bd20151f710367769ca0b331", + "prepared_review_unit_hash": "b4d3d2474892b2c331ec845fb57ffdca67449743d433224688b3ddf66e45b557", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4588, - "estimated_input_tokens": 1147, - "estimated_eval_tokens": 2550 + "serialized_chars": 4590, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4588, - "estimated_input_tokens": 1147, - "estimated_eval_tokens": 2550 + "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 target the same xdist-safe captured_info logging bug in testing_utils and share the same issue target; 45645 is the broader, more complete version that also updates CI log collection.", - "confidence": 0.95, + "summary": "Both PRs address the same xdist collision around `captured_info`/patched testing debug output, but PR 45645 is the broader superset that also updates CI log collection. PR 45639 looks like the narrower core fix; 45645 is the better canonical PR.", + "confidence": 0.92, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the best canonical representative because it implements the same xdist captured_info fix as 45639 but also updates downstream log aggregation to handle the new per-worker files.", + "canonical_pr_reason": "PR 45645 is the most complete version of the fix: it includes the xdist-safe output handling in `testing_utils.py` plus the CI/workflow changes needed to collect multiple `captured_info*.txt` files, so it subsumes PR 45639.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the strongest standalone fix: it addresses the core output-file collision and preserves CI debug log collection for the new file layout, making it the most complete solution in the cluster.", + "best_pr_reason": "PR 45645 is the best representative because it covers the same underlying xdist collision bug while also preserving downstream CI/debug-log behavior, making it the more comprehensive and merge-ready change.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying bug: xdist collisions in captured_info logging. 45645 extends the same code-path fix in testing_utils and updates CI handling, so they plausibly belong to one merged change." + "reason": "Accept: both PRs fix the same xdist collision around `captured_info` artifacts in `testing_utils.py`; 45645 extends the same fix with workflow/log-handling updates, so they are close enough to be merged as one change set." } ] }, @@ -20490,8 +20196,8 @@ }, { "label": "pull_request", - "component_index": 26, - "component_count": 28, + "component_index": 25, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20556,8 +20262,8 @@ }, { "label": "pull_request", - "component_index": 27, - "component_count": 28, + "component_index": 26, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20580,7 +20286,7 @@ "pull_request:46073|pull_request:46196", "pull_request:46154|pull_request:46196" ], - "prepared_review_unit_hash": "ec49a11704f8133fe4b50401e72aeea08aa6db605d05f68f908f52a0569b62fd", + "prepared_review_unit_hash": "c4bd2cb23e1dad30f2521a0881b22dbffcb724be34b6d1680dccf93e6cd46d50", "component_budget": { "node_count": 4, "item_count": 4, @@ -20602,48 +20308,48 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are separate model-specific PRs that all add attention-dispatch support, but they do not target the same concrete code path or merge cleanly into one change set.", - "confidence": 0.87, + "summary": "These PRs are all about adding SDPA / attn_implementation dispatch support, but each one targets a different model family and code path. They are similar feature PRs, not true duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46154 is the best representative of the cluster\u2019s common theme: a concrete implementation of `attn_implementation='sdpa'` dispatch with docs and tests, while the others are the same pattern applied to different model families.", + "canonical_pr_reason": "No single canonical PR: each PR implements the same feature pattern in a different model-specific attention path, so they should remain separate.", "best_issue_reason": null, - "best_pr_reason": "PR 46154 is the cleanest exemplar of the shared feature work: it is narrowly scoped, has supporting tests, and clearly expresses the underlying change without the extra multi-model breadth of 46196.", + "best_pr_reason": "PR 46196 is the strongest representative of the cluster because it covers two closely related model variants (SeamlessM4T and SeamlessM4T-v2) and has the broadest attention-dispatch changes.", "soft_edge_verdicts": [ { "left": "pull_request:46069", "right": "pull_request:46073", "accept": false, - "reason": "Both are attention-dispatch support PRs, but for different architectures (ShieldGemma2 vs CTRL) and different code paths; they are not the same bug/fix." + "reason": "Same broad SDPA support theme, but ShieldGemma2 and CTRL are different model implementations with different attention code paths." }, { "left": "pull_request:46069", "right": "pull_request:46154", "accept": false, - "reason": "Same broad feature idea, but ShieldGemma2 and T5 are separate model implementations with unrelated attention code paths." + "reason": "Both add attention-dispatch support, but ShieldGemma2 and T5 are unrelated architectures and the fixes are model-specific." }, { "left": "pull_request:46069", "right": "pull_request:46196", "accept": false, - "reason": "These touch different model families and different implementation paths; too broad to treat as one duplicate PR." + "reason": "Different model families and files; this is the same feature pattern, not the same concrete bug/change." }, { "left": "pull_request:46073", "right": "pull_request:46154", "accept": false, - "reason": "Both add SDPA dispatch support, but to different models (CTRL vs T5); not the same concrete change." + "reason": "CTRL and T5 each need separate dispatch plumbing; they are not the same underlying code-path fix." }, { "left": "pull_request:46073", "right": "pull_request:46196", "accept": false, - "reason": "Different model code paths and different targets (CTRL vs SeamlessM4T/v2), so they should not be merged as duplicates." + "reason": "These are separate model-specific SDPA integrations, not a single change that could plausibly be merged into one PR." }, { "left": "pull_request:46154", "right": "pull_request:46196", "accept": false, - "reason": "Both are SDPA-dispatch enablement, but they apply to distinct model stacks and code paths; too different for a duplicate merge." + "reason": "T5 and SeamlessM4T/v2 are different implementations; similarity is limited to the shared attention-dispatch feature." } ] }, @@ -20656,8 +20362,8 @@ }, { "label": "pull_request", - "component_index": 28, - "component_count": 28, + "component_index": 27, + "component_count": 27, "review_unit_index": 1, "review_unit_count": 1, "status": "reviewed", @@ -20673,40 +20379,40 @@ "soft_pairs": [ "pull_request:46152|pull_request:46207" ], - "prepared_review_unit_hash": "1f3fca4b00db0b4a4836ad2b532afc70bdde8b387fb44eaa0e0cbac795406705", + "prepared_review_unit_hash": "47ebff8889b0589ea55af2a3ba2af351eb6425439cc43afba987a7fe4e94f938", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4268, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4269, + "estimated_input_tokens": 1068, + "estimated_eval_tokens": 2392 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4268, - "estimated_input_tokens": 1067, - "estimated_eval_tokens": 2390 + "serialized_chars": 4269, + "estimated_input_tokens": 1068, + "estimated_eval_tokens": 2392 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are thematically similar because both remove source-file open() logic, but they address different code paths and are not duplicates: one is a generic import-utils helper change, the other is a modeling_utils introspection refactor.", - "confidence": 0.87, + "summary": "Both PRs aim to remove file-open/source-reading hacks, but they fix different code paths in different modules and are not mergeable as one change.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46207 is the more focused, concrete fix: it replaces one specific source-reading heuristic in modeling_utils with class introspection and includes tests.", + "canonical_pr_reason": "PR #46152 is the broader umbrella fix: it updates import-time path resolution in `import_utils.py` to avoid reading the module\u2019s own source file, which matches the cluster\u2019s shared theme most directly.", "best_issue_reason": null, - "best_pr_reason": "PR 46207 is the better representative for the cluster because it is narrowly scoped, directly tied to a single runtime behavior, and easier to evaluate independently.", + "best_pr_reason": "PR #46152 is the better cluster representative because its title and diff capture the general pattern of eliminating self-source-file reads, while #46207 is a separate, narrower introspection refactor in `modeling_utils.py`.", "soft_edge_verdicts": [ { "left": "pull_request:46152", "right": "pull_request:46207", "accept": false, - "reason": "They share a general theme (avoiding source reads), but they fix different mechanisms in different modules and are not the same underlying bug or change." + "reason": "Related pattern, but not the same underlying bug/change: one rewrites resource-path resolution in `import_utils.py`, the other changes attention-implementation capability detection in `modeling_utils.py`. They touch different subsystems and would not plausibly be one PR." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 224c0062931462b862af7f5c62837e2819fdaa98..e59289e53c918ae917cc8d6272801eba56818b75 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260526t060025z", + "analysis_id": "hybrid-model-20260526t120027z", "archived_artifacts": { - "hybrid": "snapshots/20260526T060025Z/analysis-runs/hybrid-model-20260526t060025z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260526T060025Z/analysis-runs/hybrid-model-20260526t060025z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-05-26T06:16:51Z", + "published_at": "2026-05-26T12:14:05Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260526T060025Z", + "snapshot_id": "20260526T120027Z", "variant": "hybrid" } diff --git a/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.json b/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..f5eda7d1a044a858465787cb3327a2a70f7372db --- /dev/null +++ b/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.json @@ -0,0 +1,2156 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260526T120027Z", + "generated_at": "2026-05-26T12:14:03Z", + "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-46032-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #46032.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 46032, + "canonical_pr_number": 46084, + "issue_numbers": [ + 46032 + ], + "pr_numbers": [ + 46084, + 46098 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 46084, + "right_pr_number": 46098, + "code_similarity": 0.43, + "size_similarity": 0.5, + "file_overlap": 0.667, + "area_overlap": 0.226, + "patch_similarity": 0.339, + "shared_filenames": [ + "src/transformers/models/mamba2/modeling_mamba2.py", + "tests/models/mamba2/test_modeling_mamba2.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/mamba2/modeling_mamba2.py", + "left_ranges": [ + [ + 256, + 262 + ], + [ + 303, + 353 + ], + [ + 461, + 467 + ], + [ + 471, + 491 + ], + [ + 496, + 502 + ], + [ + 599, + 609 + ] + ], + "right_ranges": [ + [ + 255, + 373 + ], + [ + 481, + 495 + ] + ] + }, + { + "filename": "tests/models/mamba2/test_modeling_mamba2.py", + "left_ranges": [ + [ + 206, + 236 + ], + [ + 285, + 322 + ] + ], + "right_ranges": [ + [ + 476, + 542 + ] + ] + } + ] + } + ] + }, + { + "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-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-46100-2", + "summary": "Cluster of 2 related pull requests linked by soft_similarity.", + "status": "open", + "confidence": 0.5, + "canonical_issue_number": null, + "canonical_pr_number": 46100, + "issue_numbers": [], + "pr_numbers": [ + 46100, + 46114 + ], + "evidence_types": [ + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 46100, + "right_pr_number": 46114, + "code_similarity": 1.0, + "size_similarity": 1.0, + "file_overlap": 1.0, + "area_overlap": 1.0, + "patch_similarity": 1.0, + "shared_filenames": [ + "tests/models/vit/test_modeling_vit.py" + ], + "shared_file_areas": [ + { + "filename": "tests/models/vit/test_modeling_vit.py", + "left_ranges": [ + [ + 216, + 226 + ] + ], + "right_ranges": [ + [ + 216, + 226 + ] + ] + } + ] + } + ] + }, + { + "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-43979-11", + "canonical_pr_number": 44007, + "duplicate_pr_numbers": [ + 43996, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "target_issue_number": 43979, + "reason": "PRs in cluster-43979-11 are treated as duplicates because they converge on issue #43979 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-44018-2", + "canonical_pr_number": 44068, + "duplicate_pr_numbers": [ + 44018 + ], + "target_issue_number": null, + "reason": "PRs in cluster-44018-2 are treated as duplicates because they share soft_similarity evidence." + }, + { + "cluster_id": "cluster-45081-3", + "canonical_pr_number": 45317, + "duplicate_pr_numbers": [ + 45086 + ], + "target_issue_number": 45081, + "reason": "PRs in cluster-45081-3 are treated as duplicates because they converge on issue #45081 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45561-3", + "canonical_pr_number": 45645, + "duplicate_pr_numbers": [ + 45639 + ], + "target_issue_number": 45561, + "reason": "PRs in cluster-45561-3 are treated as duplicates because they converge on issue #45561 with closing_reference, shared_issue_target, soft_similarity evidence." + }, + { + "cluster_id": "cluster-45647-3", + "canonical_pr_number": 45738, + "duplicate_pr_numbers": [ + 45737 + ], + "target_issue_number": 45647, + "reason": "PRs in cluster-45647-3 are treated as duplicates because they converge on issue #45647 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46032-3", + "canonical_pr_number": 46084, + "duplicate_pr_numbers": [ + 46098 + ], + "target_issue_number": 46032, + "reason": "PRs in cluster-46032-3 are treated as duplicates because they converge on issue #46032 with closing_reference, shared_issue_target evidence." + }, + { + "cluster_id": "cluster-46100-2", + "canonical_pr_number": 46100, + "duplicate_pr_numbers": [ + 46114 + ], + "target_issue_number": null, + "reason": "PRs in cluster-46100-2 are treated as duplicates because they share soft_similarity 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 13 discussion comments plus 8 inbound references.", + "score": 164.0 + }, + "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/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..bcea7d1543c411f4fd2252321cbe8d13fd1a0679 --- /dev/null +++ b/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,20427 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260526T120027Z", + "generated_at": "2026-05-26T12:14:03Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43425", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43638", + "issue:43716", + "issue:43827", + "issue:43828", + "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", + "issue:46015", + "issue:46165" + ], + "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:45901|issue:46015", + "issue:44509|issue:44512", + "issue:35545|issue:36010", + "issue:44805|issue:45823", + "issue:44918|issue:46165", + "issue:43638|issue:44805", + "issue:43638|issue:46165", + "issue:43596|issue:44805", + "issue:44805|issue:45161", + "issue:44560|issue:46165", + "issue:43596|issue:46165", + "issue:44805|issue:44918", + "issue:43572|issue:46165", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:34928|issue: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:34928|issue:41720", + "issue:43828|issue:45161", + "issue:44918|issue:45161", + "issue:43828|issue:44292", + "issue:43716|issue:45685", + "issue:43541|issue:43716", + "issue:44292|issue:44918", + "issue:43716|issue:44292", + "issue:45507|issue:45685", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43716|issue:44560", + "issue:43828|issue:45685", + "issue:44918|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:34928|issue:36010", + "issue:45237|issue:46015", + "issue:43039|issue:44805", + "issue:41762|issue:43596", + "issue:39290|issue:41720", + "issue:43901|issue:44512", + "issue:45237|issue:45901", + "issue:44841|issue:45823", + "issue:43039|issue:44560", + "issue:43975|issue:46015", + "issue:45812|issue:45920", + "issue:44863|issue:46015", + "issue:43866|issue:44863", + "issue:44292|issue:46015" + ], + "prepared_review_unit_hash": "104455d95fac3691ee93543a2b57e7f7838dca6572c1fe44d13cdd596cb0f895", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 69, + "serialized_chars": 48407, + "estimated_input_tokens": 12102, + "estimated_eval_tokens": 24460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly noisy lexical similarity across unrelated issues. The only clear duplicate pockets are the DeepSpeed ZeRO-3 zero-sized-tensor index error reports and the AutoTokenizer wrong-token-ID regression; the rest should stay separate.", + "confidence": 0.78, + "canonical_issue_reason": "issue:43596 is the best representative of the genuine duplicate pocket: it states the generic DeepSpeed ZeRO-3 zero-sized-tensor IndexError clearly, and the near-identical 43638 plus related 41762 reports orbit the same failure mode.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43596 is the strongest overall anchor for triage because it captures the core ZeRO-3 loading bug in the most general form, with the closest true duplicate nearby; most other items in the cluster are only superficially similar by error wording.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different model/config attribute error vs streaming API failure; only broad Qwen similarity." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Different hardware backends and failures (AMD GPU vs MPS histogram op); not the same bug." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Different models and symptoms: dtype mismatch in preprocessing vs AMD GPU execution failure." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Both are runtime failures, but the model, backend, and error class differ." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different MoE/runtime problems; no shared concrete code path." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Both involve Qwen/GPT-OSS style models, but the reported failures are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch is unrelated to ModernBERT ONNX export." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "ONNX export error and FineGrainedFP8 failure are separate model/tooling issues." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ModernBERT export and Qwen3 auto device mapping are unrelated." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Table QA crash and RAG docs cleanup are not the same underlying issue." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are v5 docs problems, but they cover different outdated examples and should be tracked separately." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export error vs import error for GenerationMixin; different code paths." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask/index shape mismatch is unrelated to Gemma4 device_map forward mismatch." + }, + { + "left": "issue:44918", + "right": "issue:46165", + "accept": false, + "reason": "TRl SFT embedding unpacking and resume-training/trust-remote-code are different training workflows." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Similar IndexError wording, but one is ZeRO-3/Bert loading and the other is a mask-shape mismatch." + }, + { + "left": "issue:43596", + "right": "issue:46165", + "accept": false, + "reason": "Both mention training/loading, but the actual failures are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:46165", + "accept": false, + "reason": "StableLmConfig pad token regression is unrelated to resume-training behavior." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice_preset issue has nothing to do with ONNX export." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Both involve DeepSpeed, but the collective mismatch and ZeRO-3 IndexError are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same ZeRO-3 IndexError on zero-sized tensors during Bert loading; wording differs, failure mode is the same." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger cross-entropy invocation and GPT-OSS tensor-parallel failure are different issues." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Incorrect detokenization and LoRA save/reload mismatch are separate regressions." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "NVFP4 runtime error and LoRA round-trip mismatch do not share a concrete bug." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Both are tensor-shape-related, but they involve different training stacks and error mechanisms." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "DeepSpeed weight dimension error and collective mismatch are not the same underlying bug." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Same class name, but CPU-offload incompatibility vs uninitialized-memory after meta materialization are distinct failures." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 Bert loading and GPT-OSS tensor-parallel behavior are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "Activation-checkpointing tensor-size mismatch and compute_loss signature change are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "ZeRO-3 Bert IndexError is unrelated to GPT-OSS tensor-parallel failure." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Different model families and different dtype/method errors." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Gemma3 config attribute missing and Mistral tokenizer regex patch crash are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Device placement mismatch and expensive reverse embedding lookup are different forward-path problems." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Torch version incompatibility and NVFP4 runtime error are not the same bug." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "FSDP checkpointing mismatch and Qwen3 auto device mapping failure are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs-related, but they cover different removed commands/tasks." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Mistral image dtype mismatch and Qwen3-VL video StopIteration are different model-processing failures." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Autocast dtype mismatch and MPS histogram op not implemented are different backend issues." + }, + { + "left": "issue:44918", + "right": "issue:45685", + "accept": false, + "reason": "SFT trainer embedding unpacking and MPS MoE histogram failure are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Cross-entropy being called via Liger and Phi-tiny-MoE autocast mismatch are different code paths." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "Collective mismatch in ZeRO-3 and compute_loss signature breakage are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Checkpointing/FSDP tensor mismatch and GenerationMixin import breakage are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:46015", + "accept": false, + "reason": "AMD GPU execution failure and outdated RAG docs are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger cross-entropy call and generic mask-shape IndexError are not the same bug." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": true, + "reason": "Same ZeRO-3 zero-sized-tensor IndexError during loading; Gemma3 and Bert are model-specific manifestations of the same failure." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "Missing config attribute for vLLM and CUDA assert under auto device mapping are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "TextClassificationPipeline docs and removed run-command docs are separate documentation issues." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "AMD GPU model failure and table-question-answering crash are unrelated." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are different issues." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger kernel cross-entropy path and Qwen3-VL video StopIteration are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization regression and RAG docs staleness are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same AutoTokenizer wrong-token-ID regression across different model families; titles point to the same tokenizer bug." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading and RAG docs are unrelated." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Checkpoint corruption and checkpoint loading implementation failure are different problems." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen-3 NVFP4 runtime error and RAG docs cleanup are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet. The two accepted duplicate pockets are supported by the titles and soft-pair evidence: the DeepSpeed ZeRO-3 zero-sized-tensor IndexError reports (43596/43638/41762) and the AutoTokenizer wrong-token-ID regression (45812/45920). The false soft-edge rejections are conservative and the reasons mostly distinguish different error classes, models, or code paths 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": 2, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "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:43638", + "issue:43645", + "issue:43716", + "issue:43792", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43975", + "issue:43994", + "issue:44291", + "issue:44292", + "issue:44387", + "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:43854|issue:43866", + "issue:43792|issue:45901", + "issue:41093|issue:43638", + "issue:43425|issue:45070", + "issue:43596|issue:45800", + "issue:43975|issue:45901", + "issue:43792|issue:45070", + "issue:45362|issue:45464", + "issue:44292|issue:45901", + "issue:39290|issue:45823", + "issue:45464|issue:45685", + "issue:43404|issue:43866", + "issue:43541|issue:45685", + "issue:43541|issue:43828", + "issue:44841|issue:45084", + "issue:36010|issue:39290", + "issue:45800|issue:45823", + "issue:41093|issue:41762", + "issue:43296|issue:43366", + "issue:43572|issue:45070", + "issue:43541|issue:45161", + "issue:44387|issue:45005", + "issue:43828|issue:45464", + "issue:44291|issue:45070", + "issue:43366|issue:45464", + "issue:45020|issue:45800", + "issue:43531|issue:45070", + "issue:43541|issue:44589", + "issue:41093|issue:43596", + "issue:43054|issue:43994", + "issue:45759|issue:45902", + "issue:45542|issue:45800", + "issue:39186|issue:44918", + "issue:43716|issue:44918", + "issue:40990|issue:44960", + "issue:45632|issue:45698", + "issue:43645|issue:45399", + "issue:43404|issue:45005" + ], + "prepared_review_unit_hash": "8cbde3b4a90ad0a27117859a33b3783e37858211c7834e571f5c56158a9aae4e", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 39, + "serialized_chars": 39133, + "estimated_input_tokens": 9784, + "estimated_eval_tokens": 19824 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a collection of unrelated Transformers bug reports with a few broad theme overlaps (sliding-window configs, ZeRO-3, dtype mismatches, Qwen/Gemma variants). Only one pair looks like a likely true duplicate: the two SigLIP2 issues reporting bad/garbled outputs. No PRs are present.", + "confidence": 0.83, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43994 is the best representative of the only likely duplicate area: it is broader and more directly describes the user-facing SigLIP2 failure across AutoModel/pipeline.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Both mention sliding_window, but they are different models and different failure modes (Gemma3 config field vs Qwen3-MoE behavior)." + }, + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Unrelated model-loading failure vs corrupted checkpoint report; no shared concrete bug path." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper loading/runtime issue is unrelated to table-question-answering crashes." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Both involve index/shape errors, but one is a mask mismatch and the other is a ZeRO-3 index-out-of-bounds on a different model path." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch version incompatibility and a pydantic PretrainedConfig regression are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "Both mention environment/deepspeed-ish pain, but one is a ZeRO-3 Bert index error and the other is a torch/transformers version incompatibility." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Detokenization bug in DeepSeek code is unrelated to table-question-answering task crashes." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Whisper runtime failure is unrelated to the pydantic/config regression." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both are Qwen3.5 chat-related, but one is a crash in transformers chat and the other is a streaming chat/completions API failure; not the same concrete bug." + }, + { + "left": "issue:44292", + "right": "issue:45685", + "accept": false, + "reason": "Qwen-3-8B NVFP4 runtime error is unrelated to an MPS histogram kernel missing for MoE." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Tied lm_head weights in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." + }, + { + "left": "issue:43541", + "right": "issue:45685", + "accept": false, + "reason": "Grouped_mm tracing failure in Mixtral is a different code path than the MPS Int histogram issue." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Both are dtype/runtime issues, but they affect different models and different execution paths (dynamo tracing vs autocast)." + }, + { + "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": "ImportError for GenerationMixin is unrelated to sliding-window config issues." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Version incompatibility is not the same bug as a Gemma4 device mismatch during forward." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are zero/shape index errors, but one is a mask-length mismatch and the other is a loading-time size-0 index error on Gemma3; too different to treat as the same bug." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "PaddleOCR-VL loading in vLLM and GGUF gpt-oss support are separate model-format issues." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is unrelated to the pydantic PretrainedConfig regression." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Mixtral torch-dynamo tracing failure and GPT-OSS tensor-parallel support are different problems." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 memory/OOM regression is unrelated to tied weights in translation models." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Phi-tiny-MoE autocast dtype mismatch is unrelated to Qwen3.5 streaming chat failures." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument during init_empty_weights is a different bug than the pydantic field break." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GPT-OSS GGUF support and Qwen3.5 chat streaming are different code paths." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Remote_code breakage and a torch/transformers version incompatibility are not the same underlying bug." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to the pydantic config regression." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "Grouped_mm tracing failure and missing Float8 storage support are different runtime issues." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are index errors under ZeRO-3, but they involve different models and different failure shapes." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": true, + "reason": "Both report SigLIP2 producing bad/garbled outputs; same underlying model quality/implementation problem across different entry points." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Composite Qwen config unwrapping and uninitialized rotary embedding memory are distinct Qwen3.5 issues." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard-without-TensorFlow backend error is unrelated to the torch/transformers incompatibility." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP 2-D weight error and TRL input-embedding unpacking failure are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Mistral3 image-preprocessor dtype mismatch is unrelated to Qwen3.5 embedding unpacking." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "High perplexity on gpt-oss-20b and GLM5 are different model-specific issues." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both involve remote_code/custom-module loading, but one is a leaf-directory cache collision and the other is a wrong-module load after save_pretrained; not clearly the same concrete bug." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Custom models in Jupyter and fallback to flash-attn2 being blocked are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Specific tied-weight bug in Mistral3 is not the same as the broader translation-model tied-weights report." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft pair, issue:43054 vs issue:43994, is reasonably supported by the titles/body summaries as a likely SigLIP2 duplicate area. The other verdicts are appropriately cautious and do not overclaim. No PR pair issues are present." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 3, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "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: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:44530", + "issue:44534", + "issue:44589", + "issue:44863", + "issue:44910", + "issue:44918", + "issue:44928", + "issue:45005", + "issue:45020", + "issue:45237", + "issue:45399", + "issue:45507", + "issue:45663", + "issue:45685", + "issue:45701", + "issue:45800", + "issue:45823", + "issue:45859", + "issue:45902" + ], + "soft_pairs": [ + "issue:42915|issue:45399", + "issue:44530|issue:45005", + "issue:43873|issue:45005", + "issue:44863|issue:45005", + "issue:44360|issue:44512", + "issue:44918|issue:45507", + "issue:44492|issue:44512", + "issue:43975|issue:44292", + "issue:44292|issue:45507", + "issue:43950|issue:44534", + "issue:43644|issue:43950", + "issue:39692|issue:43054", + "issue:43828|issue:44928", + "issue:36331|issue:41762", + "issue:41628|issue:45800", + "issue:43950|issue:45902", + "issue:43872|issue:44589", + "issue:43866|issue:44403", + "issue:43054|issue:45701", + "issue:41553|issue:43883", + "issue:38175|issue:43054", + "issue:45663|issue:45902", + "issue:44910|issue:45859", + "issue:41553|issue:43881", + "issue:42947|issue:44387", + "issue:43054|issue:43493", + "issue:44403|issue:44451", + "issue:43975|issue:45237", + "issue:44403|issue:44488", + "issue:43643|issue:45020", + "issue:42947|issue:43856", + "issue:41762|issue:42915", + "issue:43881|issue:43883", + "issue:45685|issue:45823" + ], + "prepared_review_unit_hash": "b813b2861095e130a357b2e6a92620d90dd67c93565871530e803f060013445c", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37575, + "estimated_input_tokens": 9394, + "estimated_eval_tokens": 19044 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous. The only clear duplicate family is the Transformers v5 non-persistent-buffer regression; most other pairs are merely same-subsystem or same-model-family similarities, not the same concrete bug.", + "confidence": 0.95, + "canonical_issue_reason": "issue:43950 is the clearest anchor for the shared v5 non-persistent-buffer regression: it states the core bug plainly, is broadly applicable, and matches the accepted duplicate follow-up issues.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43950 is the best overall representative because it captures the underlying loader/materialization regression more precisely than the narrower follow-up reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Both involve loading/runtime checks, but one is Qwen3 MoE FP8 and the other is flash-attn2 fallback gating; not the same bug." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttentionCache crash on Qwen3.5 is unrelated to tied-weights handling in translation models." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantized offloading behavior is a different code path from tied-weights/v5 translation-model issues." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint loading failure is not the same as tied-weights regressions." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "DSA indexer ReLU discussion and a docs typo are unrelated." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "TRL SFT embedding unpacking and GraniteMoEHybrid invalid method are different failures." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "Cache-strategy typo is documentation-only and unrelated to removed run-command docs." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Detokenization regression and Qwen-3-8B-NVFP4 runtime error are different concrete bugs." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "NVFP4 runtime failure and GraniteMoEHybrid invalid method do not share a concrete code path." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Both describe the same v5 non-persistent-buffer regression: buffers are corrupted/junk-filled during loading/materialization." + }, + { + "left": "issue:43643", + "right": "issue:43950", + "accept": false, + "reason": "Missing fields in trust_remote_code config output is a different symptom from non-persistent-buffer corruption." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "SigLIP2 docs example errors are not the same as poorer text embeddings." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Autocast dtype mismatch and TRL SFT embedding unpacking are different failures." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "Trainer compute_loss signature mismatch and Gemma3 ZeRO-3 loading error are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:45800", + "accept": false, + "reason": "AutoImageProcessor import failure and torch/transformers version incompatibility are distinct issues." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": true, + "reason": "The Qwen3_5Moe inv_freq uninitialized-memory report is a model-specific manifestation of the same non-persistent-buffer materialization regression." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "bitsandbytes Int8Params keyword incompatibility and Float8 storage lookup failure are different compatibility bugs." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Corrupted checkpoint and generic loading noise are not the same underlying issue." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 embedding quality and tokenization changes are not the same bug." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Bad AutoTokenizer error message for Voxtral and Molmo missing all_tied_weights_keys are unrelated." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Both are SigLIP2-related, but zero probabilities and worse text embeddings are not clearly the same concrete defect." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Gemma4 FSDP2 shared_kv_states rebuilding and Qwen3_5Moe inv_freq memory initialization are unrelated." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Qwen3.5 FlashAttention position_ids crash and CPU-offload incompatibility are different bugs." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Voxtral AutoTokenizer error messaging and glm-4v-9b loading failure are unrelated." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "LoRA gradient checkpointing ineffectiveness and int4 quantization OOM are different memory issues." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Both concern SigLIP2 quality, but one is embedding quality and the other is HF-vs-JAX discrepancy; not enough to treat as the same bug." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "Generic loading noise and a specific model load failure are not the same underlying issue." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Generic loading noise and another model-specific load failure are not the same bug." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "Remote-code config field loss is related in theme, but too different in concrete failure from broad remote_code breakage." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Both are load failures, but they point to different models and different failure modes." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "MPS histogram dtype error in MoE and Gemma4 device mismatch are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall. The summary stays conservative: 43950/44534 are clearly the same non-persistent-buffer regression, and 45902 is a plausible model-specific manifestation of that same bug. The remaining pairs are appropriately rejected as different concrete failures rather than the same issue." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "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:43782", + "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:44960", + "issue:44991", + "issue:45020", + "issue:45081", + "issue:45305", + "issue:45313", + "issue:45356", + "issue:45399", + "issue:45538", + "issue:45588", + "issue:45823", + "issue:45923" + ], + "soft_pairs": [ + "issue:43782|issue:44960", + "issue:44991|issue:45399", + "issue:43824|issue:43872", + "issue:42491|issue:44368", + "issue:45020|issue:45823", + "issue:44661|issue:45020", + "issue:43828|issue:45823", + "issue:45081|issue:45399", + "issue:45020|issue:45356", + "issue:43742|issue:45399", + "issue:43827|issue:44509", + "issue:43404|issue:43950", + "issue:43638|issue:45823", + "issue:42915|issue:45020", + "issue:43872|issue:44291", + "issue:43257|issue:43866", + "issue:43792|issue:44220", + "issue:42915|issue:43872", + "issue:41553|issue:44991", + "issue:36331|issue:41093", + "issue:43257|issue:45305", + "issue:42947|issue:45923", + "issue:43940|issue:45313", + "issue:43854|issue:44863", + "issue:44928|issue:45588", + "issue:43872|issue:45538", + "issue:41093|issue:44805", + "issue:41553|issue:44488", + "issue:44479|issue:44560", + "issue:44530|issue:44863", + "issue:24643|issue:30064", + "issue:43366|issue:43828" + ], + "prepared_review_unit_hash": "f23e573b746e9226d625d89d680de8705663d82ecb8cc7bd562bf4ef5dd4743d", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 37042, + "estimated_input_tokens": 9261, + "estimated_eval_tokens": 18778 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated regression reports; only two pairs look like true duplicates: the v5 docs/pipeline cleanup and the `_is_hf_initialized` init/loading incompatibility. Everything else is too model- or code-path-specific to merge together.", + "confidence": 0.58, + "canonical_issue_reason": "44509 is the clearest canonical issue for the docs/pipeline-removal bug: it is broader and more explicit than 43827, which is the narrower variant of the same documentation problem.", + "canonical_pr_reason": null, + "best_issue_reason": "44509 is the best issue to keep as the representative of the only clear duplicate group in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Different models and failure modes; one is Qwen3VL weight-only loading, the other is a GLM5 model request." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading failure vs flash-attn fallback gating; different bugs." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "ImportError for a missing class vs bitsandbytes Int8Params kwarg incompatibility." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "LoRA/Qwen3 MoE compatibility problem vs a gradient-checkpointing warning; not the same bug." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Broad remote_code breakage vs a Gemma4 device_map forward mismatch." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "Tokenizer mapping/model registration issue vs remote_code regressions; different code paths." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch vs device placement mismatch; not the same failure." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer regex patch crash vs flash-attn fallback logic; unrelated." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "General remote_code breakage vs a specific Kimi-K2.5 tokenizer/codec regression." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "KeyError while loading a model vs flash-attn fallback checks; unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Same v5 docs cleanup for removed summarization/translation pipeline tasks." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Tied lm_head bug vs non-persistent buffer corruption; different issues." + }, + { + "left": "issue:43638", + "right": "issue:45823", + "accept": false, + "reason": "Zero-3 index error on a Bert model vs Gemma4 PLE device mismatch." + }, + { + "left": "issue:42915", + "right": "issue:45020", + "accept": false, + "reason": "FP8/Qwen3 MoE failure vs broad remote_code regressions; not the same bug." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Both report the same `_is_hf_initialized` incompatibility during model initialization/loading." + }, + { + "left": "issue:43257", + "right": "issue:43866", + "accept": false, + "reason": "Qwen3 MoE weight conversion under accelerate+deepspeed vs a corrupted Ovis2 checkpoint." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper runtime failure vs a generic fbank feature extraction issue; too indirect to treat as duplicate." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "Different root causes: FP8 config failure vs bitsandbytes/init kwarg mismatch." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Bad AutoTokenizer error message for Voxtral vs tokenizer loading failure for a specific model." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Unexpected compute_loss kwarg vs tensor mask shape mismatch; unrelated." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "Weights not converted in Qwen3 MoE loading vs gradient averaging under DeepSpeed." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "LoRA gradient checkpointing ineffective vs missing supports_gradient_checkpointing flag; related theme but different bug." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Different Qwen variants and different ZeRO-3 load failures; not clearly the same concrete bug." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "Model-loading failure in unit tests vs NemotronH checkpoint-loading failure; different models." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "RLHF gradient explosion/NaN issue vs flash_attention.py crash on s_aux=None." + }, + { + "left": "issue:43872", + "right": "issue:45538", + "accept": false, + "reason": "bitsandbytes init kwarg incompatibility vs CLIPTokenizer max length bug." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "Both are IndexErrors, but the mask mismatch and the generic shape mismatch are different bugs." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Voxtral tokenizer error messaging vs a specific model loading failure." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": false, + "reason": "Related video-input regressions, but the impacted models and failure details are different enough to keep separate." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "Qwen3.5 PagedAttention crash vs NemotronH checkpoint loading; unrelated." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "DeepSpeed training weight-dimension error vs void segmentation-map preprocessing; unrelated." + }, + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "GGUF architecture support request vs autocast dtype mismatch; not the same issue." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The verdicts are grounded and conservative. The two accepted pairs are well supported by the packet: 43827/44509 are the same v5 docs pipeline-removal cleanup, and 43872/44291 are the same `_is_hf_initialized` loading incompatibility. The rejections generally stay within the evidence and avoid overclaiming duplicates." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "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:43720", + "issue:43824", + "issue:43950", + "issue:43957", + "issue:43994", + "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:45325", + "issue:45357", + "issue:45584", + "issue:45663", + "issue:45685", + "issue:45823", + "issue:45834", + "issue:45902" + ], + "soft_pairs": [ + "issue:45325|issue:45902", + "issue:45663|issue:45823", + "issue:43366|issue:43531", + "issue:44155|issue:45584", + "issue:43950|issue:45834", + "issue:44560|issue:45685", + "issue:43824|issue:43957", + "issue:43550|issue:44910", + "issue:43645|issue:43646", + "issue:42915|issue:43957", + "issue:43366|issue:44589", + "issue:42617|issue:43366", + "issue:44560|issue:44918", + "issue:36010|issue:42915", + "issue:44661|issue:45823", + "issue:44560|issue:44805", + "issue:43632|issue:45834", + "issue:38175|issue:43994", + "issue:43638|issue:44661", + "issue:43596|issue:44661", + "issue:44661|issue:44805", + "issue:44877|issue:45030", + "issue:43317|issue:44756", + "issue:43644|issue:44534", + "issue:38175|issue:39692", + "issue:42491|issue:43720", + "issue:41720|issue:42915", + "issue:44960|issue:44977", + "issue:44451|issue:44488", + "issue:42915|issue:44756", + "issue:45313|issue:45357", + "issue:42491|issue:45276" + ], + "prepared_review_unit_hash": "11bb5cf7a432dd1d7b343b408584ac953985b0311e3a3642602b27199accccc8", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36682, + "estimated_input_tokens": 9171, + "estimated_eval_tokens": 18598 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a loose mixed cluster: a few genuine duplicates around Transformers v5 regressions, but most items are unrelated model-specific bugs. I only accepted the clearly same-root-cause pairs.", + "confidence": 0.62, + "canonical_issue_reason": "43646 is the broadest, clearest regression report in the set: it captures the core Transformers v5 custom-model initialization breakage that best represents the small duplicate-heavy subset.", + "canonical_pr_reason": null, + "best_issue_reason": "It is the most representative issue for the cluster\u2019s regression-heavy subset and is cleaner than the highly model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Different bugs and different model families: Qwen2.5-VL rope scaling vs Qwen3.5-MoE inv_freq materialization." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4, but one is an FSDP2 shared_kv_states KeyError and the other is a device_map forward mismatch; not the same code-path." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "GGUF GPT-OSS support and Qwen3-MoE sliding_window are unrelated changes." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "AudioFlamingo3 track-leak batching bug vs Whisper empty-transcription failure; different models and failures." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent buffer corruption is unrelated to Kosmos2.5 long-OCR index errors." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3-VL video StopIteration and MoE MPS histogram dtype error are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Missing Qwen2_5_VL import and generic meta-device loading breakage are different issues." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Bamba torch.compile+SDPA crash is unrelated to Qwen3.5 flash-attention 3D position_ids handling." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Both report the same Transformers v5 custom model initialization regression; the notebook wording in 43645 is just a context variant." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "Qwen3-MoE FP8 failure and meta-device loading failures are different bugs." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF GPT-OSS support is unrelated to the Float8 storage lookup error." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "3d_parallel.py runtime failure and GGUF GPT-OSS support are unrelated." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Different Qwen3.5/Qwen3-VL audio-video embedding issues; not the same underlying bug." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "GenerationMixin import error and Qwen3-MoE FP8 failure are unrelated." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping-related add-new-model-like failure is unrelated to Gemma4 device mismatch." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Qwen3-VL video StopIteration and generic mask-shape IndexError are unrelated." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized flag regression and Kosmos2.5 OCR index error are different issues." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Both are reports of the same SigLIP2 base model producing broken/zeroed outputs through normal loading/inference paths." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "Zero3 BERT index error and tokenizer-mapping add-new-model-like failure are different code paths." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Zero3 BertModel initialization failure and tokenizer-mapping failure are unrelated." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer mapping validation and mask/tensor shape mismatch are unrelated." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both mention stricter config handling, but the model families and failure modes differ; not enough evidence of one underlying bug." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "gpu+cpu offload dequantized loading failure and Strix Halo mmap OOM are different loading/memory issues." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Same non-persistent-buffer junk regression in Transformers v5; 44534 is a later duplicate of 43644." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Same SigLIP2 family, but one is zero probabilities while the other is a doc example/model-processor/quantization error; not the same concrete bug." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "Qwen3 MoE LoRA compatibility regression and BitNet packed-weight loading bug are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Qwen3 auto device-mapping CUDA assert and Qwen3-MoE FP8 failure are distinct failures." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "GLM5 and Qwen3.5 flash-attention issues are unrelated." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": false, + "reason": "Both are BERT loading failures, but they reference different models and lack evidence of one shared root cause." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "ZeRO-3 language_model loading and save_pretrained visual-encoder key serialization are different bugs." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "Qwen3 MoE LoRA incompatibility and Gemma4 resize_token_embeddings behavior are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the accepted soft pairs are conservative. The duplicate calls around 43644/44534 and 43645/43646 are well supported, and the broader cluster description stays appropriately cautious about the many unrelated model-specific issues." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "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:43708", + "issue:43854", + "issue:43856", + "issue:43873", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44075", + "issue:44155", + "issue:44164", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44402", + "issue:44488", + "issue:44756", + "issue:44910", + "issue:44912", + "issue:44960", + "issue:44991", + "issue:45276", + "issue:45292", + "issue:45335", + "issue:45362", + "issue:45406", + "issue:45488", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:45859", + "issue:46050" + ], + "soft_pairs": [ + "issue:43708|issue:44075", + "issue:43644|issue:43645", + "issue:44488|issue:44991", + "issue:43854|issue:43931", + "issue:45488|issue:45812", + "issue:39692|issue:43994", + "issue:43257|issue:43854", + "issue:43856|issue:43873", + "issue:42491|issue:43931", + "issue:43856|issue:44387", + "issue:45362|issue:45406", + "issue:44960|issue:46050", + "issue:42491|issue:45292", + "issue:44164|issue:44291", + "issue:44910|issue:45588", + "issue:45276|issue:45335", + "issue:36010|issue:39404", + "issue:43404|issue:43454", + "issue:41720|issue:44155", + "issue:43425|issue:43957", + "issue:39692|issue:44960", + "issue:44402|issue:44488", + "issue:42947|issue:45704", + "issue:45663|issue:45759", + "issue:43873|issue:44387", + "issue:44291|issue:44756", + "issue:45663|issue:45859", + "issue:44155|issue:45823", + "issue:43335|issue:45663", + "issue:44292|issue:44912", + "issue:44387|issue:44756", + "issue:43957|issue:44756", + "issue:42947|issue:43381", + "issue:45276|issue:45292", + "issue:44164|issue:44756" + ], + "prepared_review_unit_hash": "638c884d5c32dd2d49d75e97c16b9d2d2c0c629d485f0a6671604064161cf1cc", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38298, + "estimated_input_tokens": 9575, + "estimated_eval_tokens": 19406 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are separate Transformers regressions across different models and subsystems. Only the resize_token_embeddings reports for Gemma4/generic output embeddings look like a plausible duplicate pair.", + "confidence": 0.67, + "canonical_issue_reason": "issue:45292 is the most generic description of the shared resize_token_embeddings regression; issue:45276 reads like a Gemma4-specific manifestation with extra affected modules.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the cleanest representative of the only likely duplicate pair and has the broadest, most reusable title.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Different areas: checkpoint resume step math vs SGD optimizer args being ignored." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "One is non-persistent buffer initialization junk; the other is Jupyter custom-model initialization, which is unrelated." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading failures, but for different models and likely different root causes." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Different models and symptoms: GLM-4.7 unit-test load issue vs Qwen3-VL weight-shape mismatch." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different model families and are not the same concrete bug." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "Same model family, but one is a doc-example/quantization mismatch and the other is bad inference results in AutoModel/pipeline." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different failures: Qwen3 MoE weight conversion under accelerate/deepspeed vs GLM-4.7 loading in tests." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Both touch quantization/offloading, but the reported failure modes and likely code paths differ." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Unrelated issues: Qwen3 MoE LoRA compatibility on hf5.x vs Qwen3-VL load shape mismatch." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both mention memory/OOM under quantization, but they are distinct bugs with different symptoms and causes." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Different model-specific crashes: Qwen3.5 chat vs Gemma4Processor missing _tokenizer in serve." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "GLM5 loading vs sinq save/load corruption are separate issues." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "LoRA/Qwen3 MoE compatibility is unrelated to the resize_token_embeddings regression." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Different load/save failures: extra_state handling vs init_empty_weights/_is_hf_initialized argument mismatch." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both are flash_attention crashes, but one is a packed-sequence misinterpretation on Qwen3.5 and the other is s_aux=None handling." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": false, + "reason": "Similar wording, but different model families and embedding surfaces; not clearly the same concrete bug." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to Whisper pipeline return_language behavior." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both are lm_head tying problems, but they are model-specific reports rather than the same concrete fix." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Different problems: Qwen3 auto device-map crash vs AudioFlamingo3 batched inference leakage." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Torch version incompatibility is a different issue from meta-device loading failures." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "Unrelated model families and failure modes: SigLIP2 docs/inference vs GLM5 loading." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Both involve tokenizer/model loading, but the specific failures and affected models are different." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing with PEFT LoRA is unrelated to the apex.FusedRMSNorm memory leak." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Different composite-model bugs: FSDP2 shared_kv_states handling vs text_config unwrapping in from_config." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Both are memory issues under quantization, but one is offloading behavior and the other is reserved-memory growth/OOM." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "TypeError from init_empty_weights is unrelated to mmap/OOM on Strix Halo." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Different code paths: Gemma4 FSDP2 attention state vs Qwen3 MoE rotary embedding CPU-offload compatibility." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo3 batch leakage and Gemma4 device-mapping mismatch are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Different model/config bugs: sparse-layer creation in SwitchTransformers vs Gemma4 FSDP2 attention state." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Both are quantization-related load failures, but they are for different models and likely different causes." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "OOM under int4 quantization is not the same as disabling mmap for Strix Halo memory pressure." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading regressions are distinct from the Strix Halo mmap/OOM workaround." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "Both mention gradient checkpointing, but one is PEFT LoRA ineffectiveness and the other is eval-mode incompatibility." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same underlying bug: resize_token_embeddings does not propagate to output embeddings; 45276 is just Gemma4-specific." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "Save/from_pretrained extra_state handling is unrelated to mmap-induced OOM on Strix Halo." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary stays grounded in the packet and the only accepted soft-edge pair is conservative: 45276 and 45292 both describe the same resize_token_embeddings/output_embeddings regression, with 45276 adding Gemma4-specific detail. The rest of the verdicts appropriately avoid overclaiming duplicates across distinct model families or failure modes." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36032", + "issue:39404", + "issue:43257", + "issue:43299", + "issue:43381", + "issue:43425", + "issue:43452", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43854", + "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:44683", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44877", + "issue:44898", + "issue:44912", + "issue:44918", + "issue:44928", + "issue:45084", + "issue:45216", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "issue:43854|issue:45216", + "issue:43381|issue:44928", + "issue:45684|issue:45698", + "issue:44479|issue:44877", + "issue:45588|issue:45663", + "issue:43646|issue:43708", + "issue:43645|issue:44164", + "issue:43828|issue:43957", + "issue:45292|issue:45335", + "issue:43425|issue:43828", + "issue:44402|issue:44451", + "issue:43257|issue:44863", + "issue:44589|issue:44756", + "issue:39404|issue:45701", + "issue:44683|issue:45823", + "issue:43425|issue:45823", + "issue:43611|issue:43645", + "issue:43425|issue:43716", + "issue:44530|issue:45084", + "issue:43906|issue:44530", + "issue:39404|issue:44912", + "issue:44265|issue:44898", + "issue:44479|issue:44805", + "issue:43901|issue:44509", + "issue:43994|issue:44079", + "issue:43425|issue:44918", + "issue:44589|issue:45084", + "issue:43299|issue:43931", + "issue:44164|issue:45759", + "issue:36032|issue:43452" + ], + "prepared_review_unit_hash": "22d601de719e104dac5564e55a56ad9acd71aaac6a84a0098ac2a6d7cc4f80a0", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36816, + "estimated_input_tokens": 9204, + "estimated_eval_tokens": 18664 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most of the cluster is heterogeneous. The resize_token_embeddings pair is a solid duplicate, but the Qwen3-VL MoE/loading pair is only superficially similar and should not be merged as the same bug.", + "confidence": 0.8, + "canonical_issue_reason": "issue:45292 is the broadest, most reusable statement of the resize_token_embeddings bug; issue:45335 is the t5gemma-specific manifestation of the same missing propagation.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the best representative issue in this set because it captures a concrete, generalizable bug pattern with a clear duplicate counterpart.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Both are model-loading regressions, but they involve different models and different failure modes." + }, + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Unrelated bugs: gradient checkpointing in eval mode versus Qwen3.5 RLHF NaNs from position_ids/SDPA behavior." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both touch save/load workflows, but one is about file permissions and the other about loading the wrong custom module." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Different model families and different regressions; only broad loading/config similarity." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "Different code paths: flash_attention sink handling versus Gemma4 FSDP2 shared_kv_states reconstruction." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization in notebooks is unrelated to Trainer resume_from_checkpoint max_steps math." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Both involve custom-model workflows, but notebook initialization and extra_state save/load handling are different bugs." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Different causes and symptoms: autocast dtype mismatch for Phi-tiny-MoE versus meta-device loading regressions." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "Same underlying resize_token_embeddings bug: 45335 is the t5gemma-specific instance of the output embedding not being updated." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "Torch version compatibility and a model-specific autocast dtype mismatch are not the same bug." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "The titles suggest related loading trouble, but there is not enough evidence that they share the same tokenizer/vocab mismatch." + }, + { + "left": "issue:43257", + "right": "issue:44863", + "accept": false, + "reason": "Both are load failures, but for different model families and different loading pipelines." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "A Float8 storage TypeError and an mmap OOM workaround are unrelated failure modes." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper pipeline return_language regression is unrelated to a version-induced tokenization change in general." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and Gemma4 device-map forward mismatch are different problems." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "Torch compatibility and Gemma4 PLE device mismatch are not the same underlying bug." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "base_model_prefix loading failures and notebook-defined custom model initialization failures are different issues." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch compatibility is unrelated to Mistral-3 image preprocessor/model dtype mismatch." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash is unrelated to the non-template-node compile error." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "No clear shared bug beyond broad runtime-regression similarity." + }, + { + "left": "issue:39404", + "right": "issue:44912", + "accept": false, + "reason": "Whisper pipeline language return and MXFP4 quantization fallback are different issues." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export failures with torch_compilable_check are not the same as Perceiver interpolation failures." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "Video-input regression in Qwen VL models is unrelated to the mask/tensor shape IndexError." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Both are docs issues, but they refer to different deprecations and should not be merged." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "A bad model output regression and ModelOutput key assignment behavior are distinct bugs." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch 2.10 incompatibility is unrelated to Qwen3.5 input-embedding unpacking in TRL SFT." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "Different failures: missing Float8 storage object versus compile-time rejection of non-template nodes." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "This is only supported by shared Qwen3-VL/model-family/title similarity; the packet does not show enough evidence of the same concrete bug or code path." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state save/load handling is unrelated to unwrapping text_config for composite Qwen multimodal configs." + }, + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "T5 tokenizer add_special_tokens conflict and gguf_file breaking from_pretrained are unrelated loading failures." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but one soft-edge verdict is too strong for the packet: issue:43299 vs issue:43931 is only supported by shared model family/title similarity, not clear evidence of the same underlying bug. Calling it the same loading problem overstates what the packet shows. The resize_token_embeddings pair (45292/45335) looks conservative and well-supported, but the Qwen3-VL pair should be treated as merely plausible, not accepted as a duplicate." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 8, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42175", + "issue:42915", + "issue:43317", + "issue:43421", + "issue:43452", + "issue:43482", + "issue:43541", + "issue:43582", + "issue:43596", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43856", + "issue:43883", + "issue:44164", + "issue:44206", + "issue:44220", + "issue:44451", + "issue:44464", + "issue:44479", + "issue:44488", + "issue:44530", + "issue:44661", + "issue:44671", + "issue:44683", + "issue:44737", + "issue:44756", + "issue:44811", + "issue:44849", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45584", + "issue:45859", + "issue:45987", + "issue:46178" + ], + "soft_pairs": [ + "issue:43596|issue:44683", + "issue:44671|issue:44991", + "issue:39404|issue:44671", + "issue:44991|issue:45081", + "issue:43856|issue:44756", + "issue:44164|issue:44479", + "issue:32090|issue:39404", + "issue:44737|issue:45859", + "issue:43582|issue:43632", + "issue:44661|issue:44683", + "issue:44488|issue:45081", + "issue:44488|issue:44987", + "issue:43611|issue:43644", + "issue:40990|issue:43421", + "issue:43716|issue:45084", + "issue:44206|issue:44220", + "issue:36032|issue:42915", + "issue:43883|issue:44488", + "issue:43643|issue:44530", + "issue:34567|issue:40444", + "issue:42175|issue:46178", + "issue:44811|issue:45584", + "issue:43541|issue:45084", + "issue:44451|issue:44987", + "issue:42915|issue:44530", + "issue:43644|issue:43646", + "issue:44464|issue:45987", + "issue:43452|issue:43482", + "issue:36032|issue:43645", + "issue:43317|issue:43482", + "issue:44849|issue:45325", + "issue:44987|issue:44991" + ], + "prepared_review_unit_hash": "0e3e6332b3dbb2637b15432ab57bc959f1c045393a6a94dfefd340eeff2cc456", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37234, + "estimated_input_tokens": 9309, + "estimated_eval_tokens": 18874 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of loosely related Transformers v5 regression reports. I only see one clear duplicate pair: the audio feature-extraction/fbank crash. The rest are thematically similar at best, but not the same concrete bug.", + "confidence": 0.67, + "canonical_issue_reason": "issue:44206 is the better umbrella report for the fbank/feature-extraction crash; issue:44220 looks like the same code path expressed through the helper-level failure.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44206 is the most representative and concrete issue in the only clear duplicate pair, with the broadest description of the underlying audio regression.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Different failures: deepspeed/zero3 index error in BertModel vs a torch 2.9 flex_attention compile regression." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading-related, but one is a CamemBERT prediction regression and the other is a tokenizer loading failure for a different model." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline return_language regression vs CamemBERT masked-LM output bug; unrelated code paths." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both touch tokenizer loading, but the models and failure modes differ; no evidence of the same underlying bug." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Qwen3 MoE training memory usage vs Strix Halo mmap OOM; same general theme, different causes." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "Model save/load extra_state handling vs a video-input regression; not the same bug." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error and Whisper return_language pipeline failure are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "Both involve device placement, but they affect different models and different forward paths." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "Apple Silicon allocator warmup TypeError vs _is_hf_initialized flag regression; no shared concrete bug." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer mapping/model-addition logic vs compiled flex_attention tracing; unrelated." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Both are tokenizer-loading complaints, but about different models and different breakages." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Two different model-loading failures; only a broad transformer-version theme is shared." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Both are v5 regressions, but one is base_model_prefix loading and the other is non-persistent buffer initialization." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity regression and tokenizer post-processor update bug are different issues." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Image preprocessor dtype mismatch vs template compilation TypeError; unrelated." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": true, + "reason": "Same audio feature-extraction path: one reports the unsupported center arg crash, the other reports the downstream _torch_extract_fbank_features failure." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "T5 tokenizer add_special_tokens conflict vs Qwen3Moe FineGrainedFP8Config failure; unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Missing all_tied_weights_keys on Molmo vs a model-loading regression for a different repository." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "AutoConfig trust_remote_code field loss vs Qwen3.5 cache group-type crash; different code paths." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counter not updating vs Qwen2.5-VL iterable-dataset multi-image finetuning failure." + }, + { + "left": "issue:42175", + "right": "issue:46178", + "accept": false, + "reason": "Both are backend-installation complaints, but tensorflow missing from torch extras and tensorflow_text backend mapping are distinct issues." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both involve Whisper special-token behavior, but one is batch_decode skipping tokens and the other is empty-transcription generation after alignment." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "Torch dynamo grouped_mm tracing error vs template-node compilation error; not the same bug." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Different model-loading failures; no concrete shared code-path shown." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Qwen3Moe FP8 config failure vs Qwen3.5 PagedAttentionCache group-type crash; related model family, different bugs." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 regressions, but one concerns buffer filling and the other custom model initialization." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Compiled chunked generation inconsistency vs StaticCache return-type contract bug; different failures." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Likely both concern GGUF loading, but the titles do not show the same concrete bug/change with enough confidence." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "Tokenizer special-token conflict vs Jupyter custom-model initialization breakage; unrelated." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "GPU+CPU offload dequantized-model load failure vs Qwen2.5-GGUF loading failure; not enough evidence of one bug." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Qwen3.5 hidden-states bug vs Qwen2.5-VL rope-index scaling bug; different symptoms and code paths." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "Both are loading regressions on different models, but the titles do not establish a shared underlying fixable bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays mostly conservative. The one accepted soft pair, issue:44206 vs issue:44220, is a reasonable duplicate call based on the shared audio/fbank feature-extraction path, and the other soft-edge rejections are cautious and supported by the 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": 9, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36032", + "issue:40444", + "issue:41553", + "issue:41720", + "issue:42915", + "issue:43089", + "issue:43262", + "issue:43296", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43540", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43819", + "issue:43883", + "issue:43994", + "issue:44062", + "issue:44220", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44451", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44589", + "issue:44610", + "issue:44683", + "issue:44756", + "issue:44877", + "issue:44987", + "issue:44991", + "issue:45081", + "issue:45084", + "issue:45127", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45507", + "issue:45987" + ], + "soft_pairs": [ + "issue:45084|issue:45507", + "issue:43883|issue:44991", + "issue:43262|issue:43377", + "issue:43645|issue:44534", + "issue:43089|issue:45987", + "issue:43646|issue:44534", + "issue:43404|issue:44466", + "issue:43322|issue:43540", + "issue:44403|issue:45081", + "issue:43994|issue:44610", + "issue:42915|issue:45084", + "issue:43377|issue:43819", + "issue:44265|issue:44610", + "issue:44451|issue:45081", + "issue:44987|issue:45081", + "issue:44220|issue:44683", + "issue:42915|issue:44291", + "issue:45216|issue:45357", + "issue:43632|issue:44756", + "issue:36032|issue:43742", + "issue:44062|issue:44987", + "issue:44479|issue:45325", + "issue:42915|issue:44589", + "issue:43296|issue:43531", + "issue:43404|issue:45127", + "issue:43531|issue:44877", + "issue:36032|issue:41720", + "issue:40444|issue:45325", + "issue:43611|issue:44534", + "issue:43425|issue:44403", + "issue:43632|issue:44534", + "issue:41553|issue:43582" + ], + "prepared_review_unit_hash": "20442020b040489ed9cf611af12c6355644f6b57212beab2de63cbbb7d3627e2", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36679, + "estimated_input_tokens": 9170, + "estimated_eval_tokens": 18596 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is broadly about Transformers v5-era regressions, but the individual issues span many unrelated subsystems and model families. I found no PRs, and none of the soft edges look like safe duplicates under a strict standard.", + "confidence": 0.44, + "canonical_issue_reason": "issue:43611 is the broadest, most representative loading-regression report in the set; it best fits the cluster\u2019s loose \"Transformers v5 breaks model loading\" theme, even though the cluster is heterogeneous.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43611 is the most suitable umbrella issue among these because it is concrete, widely scoped, and closest to the recurring v5 loading failures represented here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Different problems: template-node compilation vs an invalid method call in GraniteMoEHybrid." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Unrelated failures: missing all_tied_weights_keys on Molmo vs tokenizer loading for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Different domains and bugs: audio chat-template sampling rate vs MIMI encoder padding-mask mismatch." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Custom model init in notebooks is not the same as non-persistent buffers being filled with junk." + }, + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Generation performance overhead and StaticCache return-type contract are distinct issues." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Both are v5 regressions, but they concern different code paths: custom model init vs buffer serialization/init behavior." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both mention tied weights, but one is missing weight tying in model init while the other is save_pretrained serialization behavior." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Different model families and failure modes: Llava Next segfault vs Qwen3OmniMoe video-processing ValueError." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Generic loading noise is unrelated to a Mistral regex patch crash." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different models and symptoms: Siglip2 bad outputs vs OmDet-Turbo processor size mismatch." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "FineGrainedFP8 model failure is unrelated to a template-node compilation TypeError." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Both are encoder/model bugs, but the concrete failures differ: padding-mask batching vs DAC latent reconstruction." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export failure with torch_compilable_check is unrelated to OmDet processor input resolution." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Model load failure for ScandiBERT and Mistral tokenizer regex crash are different issues." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Loading physical-intelligence/fast on v5.1.0 is a different regression from the Mistral tokenizer patch crash." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction and compiled flex_attention on torch>=2.9 are separate code paths." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "FP8 config failure and init_empty_weights/_is_hf_initialized argument handling are different bugs." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": false, + "reason": "Related theme, but not clearly the same bug: general Qwen3.5 checkpoint corruption vs visual-encoder key serialization specifically." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "_is_hf_initialized breakage and mmap OOM on Strix Halo are unrelated." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "Tokenizer method conflict in T5 and a key error loading MobileLLM are not the same underlying issue." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "AddedToken special-argument conflict and a model-loading regression for physical-intelligence/fast are distinct." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": false, + "reason": "Both involve Qwen vision models, but one is a video-input regression and the other is still-image temporal position scaling." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Both touch FP8-ish areas, but one is a config/load failure and the other is a missing Float8 storage type." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL loading and Qwen3-MoE sliding_window behavior are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both involve tied weights, but one is a model class not tying lm_head while the other is LoRA merge collapse with extended vocabulary." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3-MoE sliding_window bug and granite_speech strict-config loading failure are unrelated." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "Tokenizer loading conflict in T5 and Qwen3 auto device-mapping CUDA assert are different failures." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "IterableDataset multi-image finetuning and Qwen2.5-VL rope-index scaling are different bugs." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "Base_model_prefix loading regression and non-persistent buffer junk are separate v5 issues." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch version incompatibility and generic loading noise are not the same bug." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "_is_hf_initialized handling and non-persistent buffer initialization are different code paths." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "Bad AutoTokenizer error messaging for Voxtral is unrelated to the Apple Silicon caching_allocator_warmup TypeError." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43381", + "issue:43421", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43646", + "issue:43749", + "issue:43792", + "issue:43818", + "issue:43856", + "issue:43957", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44683", + "issue:44811", + "issue:44829", + "issue:44991", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45542", + "issue:45663", + "issue:45812", + "issue:45923" + ], + "soft_pairs": [ + "issue:40990|issue:44811", + "issue:44079|issue:45663", + "issue:43792|issue:44589", + "issue:33666|issue:40444", + "issue:44683|issue:45020", + "issue:44403|issue:44991", + "issue:43421|issue:44568", + "issue:43381|issue:45923", + "issue:35707|issue:40444", + "issue:44462|issue:45812", + "issue:44291|issue:44589", + "issue:43482|issue:45538", + "issue:42175|issue:45542", + "issue:44829|issue:45479", + "issue:43317|issue:43856", + "issue:43482|issue:43525", + "issue:44079|issue:44534", + "issue:43818|issue:45357", + "issue:43531|issue:44291", + "issue:43377|issue:45491", + "issue:41720|issue:45081", + "issue:40444|issue:45381", + "issue:44265|issue:45072", + "issue:43644|issue:44079", + "issue:43749|issue:44079", + "issue:42175|issue:43296", + "issue:45325|issue:45381", + "issue:43632|issue:45072", + "issue:43611|issue:43646", + "issue:43792|issue:43957" + ], + "prepared_review_unit_hash": "60f98f8f3e0f69d4a3c5c4b0ea97015f5970f78272763cf723dbd97719939ac1", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36902, + "estimated_input_tokens": 9226, + "estimated_eval_tokens": 18708 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous; most soft pairs are related by broad subsystem/topic only, not the same bug. Two pairs do look like genuine duplicates: the AutoTokenizer/tokenizer.json regression and the Qwen2.5-VL position-id regression. I\u2019d use the tokenizer-loading issue as the main representative.", + "confidence": 0.61, + "canonical_issue_reason": "issue:44462 is the broadest concrete bug here: AutoTokenizer ignoring tokenizer.json. It also cleanly subsumes the Granite-specific symptom in issue:45812, making it the best umbrella issue in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44462 is the best representative because it is a clear, root-cause tokenizer-loading regression rather than a model-specific symptom, and it matches the strongest duplicate pair in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Different problems and code paths: perplexity evaluation vs Whisper batch_decode special-token handling." + }, + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "ModelOutput key assignment bug vs Gemma-4 FSDP shared_kv_states rebuild; unrelated." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure and Float8 storage lookup error are different failures." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Both are Qwen VL training issues, but the failures are different: multi-GPU training vs iterable dataset with multiple images." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and remote_code model loading regressions are unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Generic loading noise vs tokenizer loading failure; not the same underlying bug." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Both involve tokenizer special tokens, but one is runtime post-processor refresh and the other is BOS/EOS insertion for a specific tokenizer." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Eval-mode gradient checkpointing misuse is not the same as a missing supports_gradient_checkpointing flag." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen VL finetuning." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": true, + "reason": "Both point to the same AutoTokenizer loading regression; Granite wrong token IDs plausibly come from ignoring tokenizer.json." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and missing Float8 storage object are different init/load failures." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "Qwen2.5-GGUF loading failure and CLIPTokenizer model_max_length are unrelated tokenizer/model-loading bugs." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both touch TensorFlow packaging, but one is missing backend in the torch extra and the other is a tf-backend detection/runtime issue." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Flash-attention-3 training degeneration and single-label num_labels=1 zero-loss are different classification bugs." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Device-map offload loading failure and Qwen3 MoE memory inefficiency are distinct issues." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Different model/config failures: GGUF loading vs missing pad_token_id on Llama4Config." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "ModelOutput key assignment and non-persistent buffer junk are separate core regressions." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Video-LLaVA tower weights/temporal attention issue is unrelated to Qwen3.5 save_pretrained key regression." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE and init_empty_weights argument handling are unrelated." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Both concern batching/padding, but they are different model-specific bugs with different symptoms." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Qwen3 auto device-map assert and Mistral tokenizer regex patch crash are unrelated." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL, but one is a finetuning/dataset failure and the other is a video position-id bug." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export/torch_compilable_check failure and bfloat16 dtype mismatch are unrelated." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Non-persistent buffer junk and ModelOutput key assignment are different bugs." + }, + { + "left": "issue:43749", + "right": "issue:44079", + "accept": false, + "reason": "FSDP CPU RAM efficient loading regression is unrelated to ModelOutput key assignment." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "TensorFlow extra/backend packaging and PaddleOCR-VL model loading are different issues." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": true, + "reason": "Both describe the same Qwen2.5-VL position-id calculation regression affecting image/video paths in the same area of code." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "_is_hf_initialized regression and bfloat16 dtype mismatch are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are transformers v5 regressions, but base_model_prefix loading and custom model initialization are not clearly the same bug." + }, + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "Whisper can\u2019t run and torch.device('meta') loading regressions are different failures." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet overall, and the two accepted soft pairs are reasonably conservative based on the issue titles and subsystem overlap. The representative/canonical choice of issue:44462 is also defensible for this cluster." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:34634", + "issue:35707", + "issue:39692", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43262", + "issue:43296", + "issue:43317", + "issue:43322", + "issue:43377", + "issue:43454", + "issue:43482", + "issue:43526", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43701", + "issue:43746", + "issue:43819", + "issue:43873", + "issue:43881", + "issue:43883", + "issue:43994", + "issue:44186", + "issue:44265", + "issue:44291", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44568", + "issue:44610", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45584", + "issue:45701", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45834", + "issue:45920", + "issue:45923", + "issue:46050" + ], + "soft_pairs": [ + "issue:45325|issue:45834", + "issue:43994|issue:44898", + "issue:43454|issue:44466", + "issue:42915|issue:43482", + "issue:43296|issue:45800", + "issue:43701|issue:45325", + "issue:44479|issue:45823", + "issue:44291|issue:44877", + "issue:43881|issue:44488", + "issue:43606|issue:44265", + "issue:43873|issue:45701", + "issue:43262|issue:43746", + "issue:40990|issue:45584", + "issue:34634|issue:35707", + "issue:44568|issue:45920", + "issue:35707|issue:41720", + "issue:44186|issue:44265", + "issue:43606|issue:43819", + "issue:41720|issue:45823", + "issue:43454|issue:43550", + "issue:39692|issue:40444", + "issue:43377|issue:44610", + "issue:43454|issue:45072", + "issue:43572|issue:45759", + "issue:43257|issue:45923", + "issue:43317|issue:43873", + "issue:43377|issue:44898", + "issue:40990|issue:42947", + "issue:43526|issue:43883", + "issue:43322|issue:46050" + ], + "prepared_review_unit_hash": "8fda98609b42870fc40d8e0a1655411c72060335a3e06f3031b451b7ffc18eed", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36832, + "estimated_input_tokens": 9208, + "estimated_eval_tokens": 18672 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a set of largely unrelated bug reports across different models and subsystems. The soft-similarity links are mostly superficial (e.g., both mention loading, offload, device maps, or tokenization) but do not point to the same concrete bug or mergeable change set.", + "confidence": 0.96, + "canonical_issue_reason": "No clear canonical issue: the items span different models, APIs, and failure modes, so picking one as the duplicate target would be misleading.", + "canonical_pr_reason": null, + "best_issue_reason": "No single best issue for deduplication; the cluster is heterogeneous, so none is a safe representative anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both are multimodal indexing bugs, but they affect different models and different code paths (Qwen2.5-VL RoPE temporal positions vs Kosmos2.5 OCR long-input indexing)." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different models and failures: SigLIP2 produces bad results via AutoModel/pipeline, while Perceiver fails on non-default image resolution and interpolation." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both involve tied lm_head weights, but one is a model/config tie failure and the other is a serialization inconsistency by device; not the same concrete bug." + }, + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Different models and loading paths: Qwen3Moe FP8 configuration failure vs Qwen2.5-GGUF loading failure under v5." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "Unrelated failures: PaddleOCR-VL loading in vLLM versus a torch/transformers version incompatibility." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL RoPE/position-id scaling." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Different regressions in different models: video-input handling for Qwen VL variants versus Gemma4 PLE device mismatch during forward." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "Different problems: init_empty_weights argument handling versus strict config rejection for granite_speech." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Both are load failures, but for unrelated models and likely unrelated causes." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch for Bark is unrelated to torch.export failure on torch_compilable_check models." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Offloading/quantization behavior is unrelated to tokenization changes." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio chat-template sampling-rate defaulting has no meaningful overlap with PEFT checkpoint loading." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "Different domains: perplexity on a language model versus Whisper generation failing on empty transcription." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "Bark voice preset bug and progressive generation with inputs_embeds/past_key_values are separate issues." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "Both are tokenizer regressions, but they describe different symptoms and model families; no shared concrete bug is evident." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation with inputs_embeds is unrelated to Qwen3 device-map CUDA assertion." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "Tokenizer NER/padding crashes and torch.export torch_compilable_check failures are separate code paths." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Both are audio-model-related, but one is offload device mismatch and the other is DAC from_latents vs forward mismatch due to missing STE." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Both mention device mapping, but the models and failure mechanisms differ (CUDA assert on Qwen3 vs Gemma4 PLE forward mismatch)." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "Tied-weight lm_head behavior and Bamba torch.compile SDPA failure are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "Different multimodal workflows and different bugs: SigLIP2 docs/model mismatch versus Qwen2.5-VL IterableDataset multi-image finetuning failure." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI padding-mask batching bug and OmDet-Turbo processor/model input-size mismatch are unrelated." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "Different issues around model behavior: tied lm_head weights versus dtype mismatches in bfloat16 inference." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "A missing pad_token_idx in StableLmConfig is unrelated to AutoModelForCausalLM.from_config not unwrapping text_config." + }, + { + "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:43317", + "right": "issue:43873", + "accept": false, + "reason": "Both involve offload/quantization, but one is loading dequantized models with device_map=auto and the other is a broader offloading expectation issue; not clearly the same bug." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Segmentation fault on Llava Next loading is unrelated to sinq quantized model save/load breakage." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36010", + "issue:36331", + "issue:40990", + "issue:41720", + "issue:42617", + "issue:42915", + "issue:43295", + "issue:43296", + "issue:43317", + "issue:43322", + "issue:43404", + "issue:43450", + "issue:43454", + "issue:43526", + "issue:43550", + "issue:43577", + "issue:43606", + "issue:43708", + "issue:43720", + "issue:43749", + "issue:43873", + "issue:43881", + "issue:44008", + "issue:44186", + "issue:44387", + "issue:44423", + "issue:44466", + "issue:44479", + "issue:44534", + "issue:44610", + "issue:44671", + "issue:44743", + "issue:44756", + "issue:44877", + "issue:44898", + "issue:44918", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45823" + ], + "soft_pairs": [ + "issue:43881|issue:44991", + "issue:43454|issue:43720", + "issue:41720|issue:43322", + "issue:43296|issue:44877", + "issue:43749|issue:44534", + "issue:44466|issue:45357", + "issue:41720|issue:44977", + "issue:42617|issue:42915", + "issue:40990|issue:45701", + "issue:44387|issue:45704", + "issue:44008|issue:45823", + "issue:36331|issue:42617", + "issue:44186|issue:45072", + "issue:43577|issue:43708", + "issue:44877|issue:44991", + "issue:36331|issue:42915", + "issue:43404|issue:45357", + "issue:43450|issue:45701", + "issue:43873|issue:44756", + "issue:43550|issue:43720", + "issue:44423|issue:45406", + "issue:41720|issue:42617", + "issue:45632|issue:45684", + "issue:44479|issue:44918", + "issue:43550|issue:43606", + "issue:44610|issue:44898", + "issue:44186|issue:44610", + "issue:42915|issue:43317", + "issue:43295|issue:45701", + "issue:43526|issue:43881", + "issue:41720|issue:45161", + "issue:36010|issue:42617", + "issue:44671|issue:44743" + ], + "prepared_review_unit_hash": "b24425072f9083352865720ca9fa6c385311f89b6f8928a95f8b03e9c9687baa", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 37126, + "estimated_input_tokens": 9282, + "estimated_eval_tokens": 18820 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: the items cover unrelated loading, processor, quantization, multimodal, and trainer regressions. There is no strong duplicate group; at best, a few pairs share a broad subsystem, but not the same concrete bug.", + "confidence": 0.27, + "canonical_issue_reason": "issue:43873 is the broadest and most central open report in the set, covering offloading/quantization behavior that loosely overlaps several memory/offload-related items, so it is the best anchor if a canonical issue must be chosen.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the most generally framed and still active issue among the candidates, making it the least arbitrary representative of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Both are model-loading failures, but one is about a specific GLM vision model and the other about tokenizer loading for a different checkpoint; no shared code-path bug is evident." + }, + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Both mention weights during loading, but one is about tying lm_head weights and the other about unpacking packed BitNet weights; different mechanisms and fixes." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Both are inference/loading failures, but the reported symptoms and model paths differ (Qwen3 auto device mapping vs Llava Next segfault)." + }, + { + "left": "issue:43296", + "right": "issue:44877", + "accept": false, + "reason": "These are both load failures, but one is a vLLM/transformers compatibility issue and the other is strict config validation; different root causes." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "Both touch memory/offload behavior, but one is an FSDP CPU RAM-efficient loading regression and the other is a v5 non-persistent buffer serialization problem." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both involve saving/serialization regressions, but one is tied weights serialization and the other is incorrect visual encoder key saving; not the same bug." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Both involve Qwen3.5-family models, but one is a device-mapping crash and the other is a flash-attention generation regression." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "Unrelated failures: a 3d_parallel.py runtime problem versus a Qwen3Moe FineGrainedFP8Config issue." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "Tokenizer/tokenization behavior could affect perplexity, but the reports do not describe the same underlying defect or code path." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Both are memory-related, but one is int4 quantization causing extra reserved CUDA memory and the other is an apex RMSNorm memory leak." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma-family model bugs, but one is a variable-name collision in forward and the other is a device-mapping mismatch in a different model." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "A trainer API signature change is unrelated to the 3d_parallel.py failure." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Different subsystems entirely: tokenizer padding/truncation versus bfloat16 dtype mismatches in model inference." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "One is a dtype-loading bug in BLIP2, the other is a resume/max_steps calculation bug in Trainer; no shared code path." + }, + { + "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; not the same defect." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Trainer loss-signature handling and FP8 model loading are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Both concern multimodal model internals, but one is weight tying and the other is save_pretrained key serialization; different bugs." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video processor batching shape issues are unrelated to the tokenization regression claim." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Both are memory/offload themed, but one is quantization offloading and the other is mmap-related OOM avoidance; distinct root causes." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "One is a torch.compile/SDPA failure in Bamba, the other is an accelerate loading issue for BitNet packed weights." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both are serve-time multimodal failures, but the specific crashes differ and point to different processor/code-path issues." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Cache-path collision and read-only permission propagation are unrelated save/load concerns." + }, + { + "left": "issue:44479", + "right": "issue:44918", + "accept": false, + "reason": "Both mention Qwen3.5, but one is a video-input regression and the other is TRL SFT embedding unpacking; not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Different failures: torch.compile with SDPA versus CPU offload device mismatch." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are image-size/preprocessing issues, but they involve different models and different expected-input mismatches." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer padding/NER crashes and image processor resolution mismatch are different preprocessing bugs." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "FP8 model failure and dequantized-model offload failure are related only at a very high level; they are not the same code-path problem." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "A regression around processor.tokenizer/image handling is not the same as a general tokenization-change report." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "Different components and symptoms: BEiT label reduction versus GLM model loading." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Tensor-parallel GPT-OSS MoE issues and Qwen3 auto device mapping failures are different model/runtime problems." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "Importing GenerationMixin is unrelated to a 3d_parallel.py execution failure." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "CamemBERT masked-LM prediction errors and Qwen3.5 recurrent-state cache resets are unrelated model-specific 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": 13, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:29159", + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36032", + "issue:36331", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43316", + "issue:43322", + "issue:43329", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43606", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43873", + "issue:43927", + "issue:44186", + "issue:44361", + "issue:44451", + "issue:44492", + "issue:44509", + "issue:44514", + "issue:44545", + "issue:44625", + "issue:44898", + "issue:44913", + "issue:44991", + "issue:45059", + "issue:45072", + "issue:45081", + "issue:45406", + "issue:45464", + "issue:45479", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45812" + ], + "soft_pairs": [ + "issue:45081|issue:45812", + "issue:43531|issue:43572", + "issue:43316|issue:44913", + "issue:43540|issue:43782", + "issue:42915|issue:43322", + "issue:44186|issue:44361", + "issue:43421|issue:45812", + "issue:36331|issue:41720", + "issue:29159|issue:36032", + "issue:43388|issue:45059", + "issue:45406|issue:45464", + "issue:33357|issue:35707", + "issue:30064|issue:34634", + "issue:43606|issue:43720", + "issue:43550|issue:43653", + "issue:41720|issue:43746", + "issue:43873|issue:45704", + "issue:43421|issue:45701", + "issue:44492|issue:44509", + "issue:43606|issue:43653", + "issue:43531|issue:45759", + "issue:34634|issue:41720", + "issue:42490|issue:43329", + "issue:34567|issue:35707", + "issue:44514|issue:44545", + "issue:44625|issue:45479", + "issue:43421|issue:43927", + "issue:44898|issue:45072", + "issue:44451|issue:44991" + ], + "prepared_review_unit_hash": "02752ac3460fa627f400bc3317b32f9cf468b4c3d7e1138853795ac26a2485fe", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36630, + "estimated_input_tokens": 9158, + "estimated_eval_tokens": 18572 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly heterogeneous; the only clear duplicate is the Qwen2_5_VLProcessor batched `padding=False` crash (44514/44545). All other soft pairs share a subsystem at most, but not the same concrete bug.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 44514 is the cleaner anchor for the only true duplicate pair here: it matches 44545 exactly on the same `Qwen2_5_VLProcessor.apply_chat_template` batched `padding=False` crash.", + "canonical_pr_reason": null, + "best_issue_reason": "44514 is the most actionable and clearly duplicated issue in the set; it best represents the one confirmed bug cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45081", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer-related, but they describe different failures: Mistral regex patching crash vs Granite token ID regression." + }, + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Different model/config bugs: Qwen3-MoE sliding window handling vs StableLmConfig missing `pad_token_idx`." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Both are config discrepancies, but for different fields and models (`Gemma3TextConfig` API vs GPTNeoX `rotary_pct` reload behavior)." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "Different code paths: Qwen3OmniMoe video input processing vs Qwen3VL `from_pretrained` weight-only loading." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "Different failures and model families: FineGrainedFP8Config issue vs Llava Next segmentation fault on load." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Both hit tokenizers, but the bugs are different: NER/batched padding crash vs MLuke task-time AttributeError." + }, + { + "left": "issue:43421", + "right": "issue:45812", + "accept": false, + "reason": "Runtime special-token post-processor sync bug is not the same as Granite tokenizer returning wrong token IDs." + }, + { + "left": "issue:36331", + "right": "issue:41720", + "accept": false, + "reason": "Trainer signature mismatch and Qwen3 device-mapping CUDA assert are unrelated." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "Both involve tokenizers, but one is slow/fast inconsistency and the other is a T5 special-token name conflict on load." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Metric-gathering label truncation is unrelated to SAM3 text/bounding-box processing." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "Different surfaces and failures: missing `_tokenizer` in `Gemma4Processor` vs Qwen3.5 streaming chat/completions failure." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "A MacOS CLIP bus error is not the same bug as progressive generation with `inputs_embeds`/`past_key_values`." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Image processor segmentation-map handling and Bark voice_preset behavior are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "Both touch loading/offload paths, but one is CPU-offload device mismatch and the other is BitNet packed-weight unpacking." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "Different model/paths: torch.compile + SDPA failure vs tokenizer special-token registration bug." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "CUDA assert during auto device mapping is unrelated to GraniteSpeech PEFT local-checkpoint loading." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantized offloading behavior and T5 Apex RMSNorm memory leak are separate issues." + }, + { + "left": "issue:43421", + "right": "issue:45701", + "accept": false, + "reason": "Special-token post-processor refresh is not the same as a general tokenization regression across versions." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A cache-strategy typo and stale docs about removed pipeline tasks are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Same project area, but different concrete bugs: CPU-offload mismatch vs BigBird tokenizer mask token registration." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Both involve Qwen3.5/Qwen3-MoE config handling, but one is sliding-window behavior and the other is `from_config` text_config unwrapping." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice preset and Qwen3 auto device mapping are unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "Import-path slash warning is unrelated to undefined video-token helper variables in multimodal processors." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "Trainer token-count tracking and progressive generation are different bugs." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "Same exact bug report and same crash condition: `Qwen2_5_VLProcessor.apply_chat_template` on batched input with `padding=False`." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Both concern classification labels, but one is config propagation and the other is a degenerate loss formulation with `num_labels=1`." + }, + { + "left": "issue:43421", + "right": "issue:43927", + "accept": false, + "reason": "Runtime special-token post-processor updates are a different bug from DiaConfig losing custom token IDs after save/load." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Image-resolution interpolation failure is unrelated to dtype mismatch in bf16 inference." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but they affect different models and do not look like the same underlying defect." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet: the 44514/44545 pair is an exact title-level duplicate with the same `Qwen2_5_VLProcessor.apply_chat_template` batched `padding=False` crash, and the other soft pairs are handled conservatively without overstating evidence. No mergeability issues are apparent." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:30064", + "issue:36010", + "issue:39692", + "issue:40444", + "issue:41720", + "issue:42886", + "issue:43381", + "issue:43421", + "issue:43531", + "issue:43596", + "issue:43645", + "issue:43653", + "issue:43720", + "issue:43746", + "issue:43782", + "issue:43819", + "issue:43844", + "issue:43873", + "issue:43994", + "issue:44062", + "issue:44077", + "issue:44186", + "issue:44220", + "issue:44368", + "issue:44451", + "issue:44589", + "issue:44610", + "issue:44625", + "issue:44661", + "issue:44671", + "issue:44843", + "issue:44918", + "issue:44936", + "issue:44987", + "issue:45042", + "issue:45081", + "issue:45356", + "issue:45357", + "issue:45381", + "issue:45588", + "issue:45663", + "issue:45704", + "issue:45759", + "issue:45859", + "issue:45907", + "issue:45923", + "issue:46050" + ], + "soft_pairs": [ + "issue:45759|issue:45859", + "issue:44077|issue:44625", + "issue:43531|issue:44589", + "issue:43782|issue:45381", + "issue:44987|issue:45042", + "issue:45588|issue:45759", + "issue:44062|issue:44451", + "issue:44368|issue:45357", + "issue:39692|issue:44610", + "issue:43746|issue:46050", + "issue:45704|issue:45923", + "issue:44918|issue:44936", + "issue:43421|issue:43653", + "issue:45663|issue:45923", + "issue:43381|issue:43844", + "issue:42886|issue:44843", + "issue:45704|issue:45859", + "issue:43531|issue:43645", + "issue:41720|issue:45704", + "issue:43421|issue:44186", + "issue:43994|issue:44671", + "issue:43746|issue:43873", + "issue:39692|issue:43873", + "issue:45081|issue:45356", + "issue:43596|issue:44220", + "issue:40444|issue:43746", + "issue:45356|issue:45907", + "issue:30064|issue:36010", + "issue:43720|issue:43819", + "issue:44220|issue:44661", + "issue:43653|issue:43720" + ], + "prepared_review_unit_hash": "4a9c0e53199f851dce4bf3c287d9888061b53b55ea916e3e47ceb13063eb368d", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36774, + "estimated_input_tokens": 9194, + "estimated_eval_tokens": 18644 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous and only one soft edge looks like a true duplicate: the offline tokenizer/cache regression around HF_HUB_OFFLINE. The rest are distinct bugs across unrelated model, tokenizer, training, and quantization paths.", + "confidence": 0.31, + "canonical_issue_reason": "Issue 44843 is the clearest root-cause-style report in the cluster and appears to subsume the offline tokenizer/cache failure described in 42886 via the _patch_mistral_regex/model_info path.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44843 is the strongest representative because it isolates a specific failing code path rather than only a symptom, but the overall cluster is not a single coherent duplicate set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Different bugs: composite-config unwrapping in from_config vs CPU-offload incompatibility in rotary embedding." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "Unrelated model/config issues; one is a patchtsmixer post_init policy change, the other is Qwen3.5 num_labels propagation." + }, + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Different subsystems and failures: Qwen3-MoE sliding window vs missing Float8 storage type." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen-VL related, but one is weight_only loading and the other is video vision_position_ids; not the same bug." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "Different issues: model loading regression for one package vs PIL image processors wrongly requiring torchvision." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Flash-attention sink-less model crash is unrelated to Qwen composite config unwrapping." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "Different failure modes: AddedToken special-arg TypeError vs a model load failure for ScandiBERT." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "One is a tie_word_embeddings warning, the other is incorrect visual-encoder key saving; not the same code path." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Both mention vision models, but one is a SigLIP2 doc/example mismatch and the other is an OmDet-Turbo processor size bug." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Different checkpoint/loading problems: PEFT adapter local load vs sinq quantized save/load corruption." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "T5 RMSNorm memory leak is unrelated to Nemotron gradient-checkpointing flag support." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Different training failures: Qwen3.5 embedding unpacking in TRL SFT vs trainer.evaluate() after trainer.train()." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Related to special tokens, but not the same bug: runtime post-processor updates vs BigBirdTokenizer mask-token registration/decode output." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Gemma-4 FSDP2 shared_kv_states KeyError is unrelated to gradient-checkpointing flag metadata." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Eval-mode gradient checkpointing limitation is unrelated to a ZeRO-3 gradient blow-up." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": true, + "reason": "Same underlying offline tokenizer/cache regression; 44843 identifies the culprit (_patch_mistral_regex calling model_info unconditionally) behind 42886's HF_HUB_OFFLINE failure." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "Different issues: T5 RMSNorm memory leak vs Qwen3_5MoeTextRotaryEmbedding CPU-offload incompatibility." + }, + { + "left": "issue:43531", + "right": "issue:43645", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to Jupyter custom-model initialization in Transformers 5.0.0." + }, + { + "left": "issue:41720", + "right": "issue:45704", + "accept": false, + "reason": "CUDA assert from auto device mapping is unrelated to the T5 apex RMSNorm memory leak." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Special-token post-processor behavior is unrelated to LayoutLMv2Tokenizer NER/padding crashes." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Different model families and bugs: SigLIP2 inference regression vs CamemBERT masked-LM prediction regression." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Local checkpoint loading with PEFT adapters is unrelated to quantization offloading behavior." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "Doc/example errors for SigLIP2 are unrelated to quantization/offloading behavior." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both involve tokenizer regex fixes, but one is a hard crash loading Mistral tokens and the other is a codec-handling regression in Kimi-K2.5; not clearly the same bug." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "Different low-level runtime errors: DeepSpeed ZeRO-3/BertModel index error vs _torch_extract_fbank_features issue." + }, + { + "left": "issue:40444", + "right": "issue:43746", + "accept": false, + "reason": "Multimodal iterable dataset fine-tuning failure is unrelated to PEFT local checkpoint loading." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer codec/regex regression is unrelated to list_repo_files not forwarding revision." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Completely different topics: void segmentation map processing vs GenerationMixin import error." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "Different numerical/model bugs: packed-weight loading vs DAC.from_latents missing STE mismatch." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio feature extraction bug is unrelated to add-new-model-like failing for TOKENIZER_MAPPING_NAMES." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "BigBird special-token decode issue is unrelated to BitNet packed-weight loading." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative overall. The only likely duplicate edge called out (42886 \u2194 44843) is supported by the issue titles and the shared _patch_mistral_regex/model_info path, while the rest of the soft-edge rejections are plausibly distinct bugs. No clear overstatement beyond the cautious wording already used." + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:30064", + "issue:36331", + "issue:39692", + "issue:41720", + "issue:43295", + "issue:43322", + "issue:43335", + "issue:43388", + "issue:43421", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43577", + "issue:43618", + "issue:43643", + "issue:43653", + "issue:43697", + "issue:43716", + "issue:43756", + "issue:43782", + "issue:43818", + "issue:43819", + "issue:43950", + "issue:44008", + "issue:44062", + "issue:44186", + "issue:44442", + "issue:44448", + "issue:44488", + "issue:44671", + "issue:44811", + "issue:44821", + "issue:44843", + "issue:44877", + "issue:44898", + "issue:44977", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45183", + "issue:45381", + "issue:45479", + "issue:45584", + "issue:45685", + "issue:45834", + "issue:45987", + "issue:46032" + ], + "soft_pairs": [ + "issue:43540|issue:44008", + "issue:43618|issue:43697", + "issue:43550|issue:45072", + "issue:45183|issue:45479", + "issue:43295|issue:45834", + "issue:43550|issue:43819", + "issue:43388|issue:46032", + "issue:44062|issue:44488", + "issue:43716|issue:44008", + "issue:43335|issue:43756", + "issue:43322|issue:45685", + "issue:44186|issue:44898", + "issue:43653|issue:43819", + "issue:43531|issue:43643", + "issue:43643|issue:44877", + "issue:43950|issue:45987", + "issue:44843|issue:45081", + "issue:30064|issue:36331", + "issue:43540|issue:45381", + "issue:44448|issue:44671", + "issue:43421|issue:44811", + "issue:41720|issue:44991", + "issue:39692|issue:45834", + "issue:43577|issue:43818", + "issue:43782|issue:44977", + "issue:45584|issue:45834", + "issue:44821|issue:45084", + "issue:44186|issue:44442" + ], + "prepared_review_unit_hash": "53ec5b1d296865e1d3fa27dd7e381108dd19a9c322ce8055d21c4186ff074185", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36507, + "estimated_input_tokens": 9127, + "estimated_eval_tokens": 18510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the soft pairs share broad Transformers themes but not the same concrete bug or fix path, so I would not deduplicate any of them.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue stands out because the cluster mixes unrelated regressions across models, tokenizers, processors, cache logic, and training utilities; if forced to pick one anchor, issue 44843 is the clearest core-loading regression, but it does not represent the rest.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44843 is the most concrete and broadly actionable regression in the set, with a clear failure mode and code path, though it is not a true representative of the whole cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different models and root causes: Qwen3OmniMoe video-processing ValueError vs Gemma 3n audio_mel_mask attribute collision." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Both are model output regressions, but one is missing CLIP attentions and the other is RTDetrV2 producing different outputs; not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Distinct code paths and failures: Bamba torch.compile/SDPA issue vs SwitchTransformers/TimmWrapperModel dtype mismatch." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "One is input validation/error-message logic; the other is zero-loss behavior in single-label classification. Different bugs." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Both involve model loading/processing, but the regression in processor.tokenizer access is unrelated to Kosmos2.5's long-OCR index error." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Unrelated failures: torch.compile with SDPA in Bamba vs DAC.from_latents not matching forward pass due to missing STE." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Different areas: gather_for_metrics last-batch label handling vs Mamba2Mixer cache correctness for seq_len > 1." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "Tokenizer construction bug with AddedToken kwargs vs failure to load a specific model tokenizer; no shared root cause." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Different models and modalities: Blip2 image preprocessor dtype mismatch vs Gemma 3n variable-name collision in audio forward pass." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Different model/config bugs: sparse layer creation with zero layers vs Smollm3 RoPE layer count mismatch." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Llava Next loading segfault and MPS histogram-not-implemented are separate platform/runtime issues." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer crash on NER padding/truncation vs Perceiver image-classification failure at non-default resolution." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug vs DAC.from_latents/forward mismatch; different subsystems." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to AutoConfig trust_remote_code missing fields." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both are config-loading issues, but one is missing fields from trust_remote_code and the other is strict config rejecting granite_speech; different causes." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "Non-persistent buffer corruption in from_pretrained is unrelated to StaticCache.get_seq_length returning the wrong type." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Both touch Mistral regex patching, but one is an offline hub call in _patch_mistral_regex while the other is an AttributeError on backend_tokenizer; different failure points." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image segmentation-map processing and CustomTrainer.compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Same broad Qwen multimodal area, but different models and bugs: Qwen3OmniMoe video input handling vs qwen2.5-vl vision_position_ids." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Different models and regressions: Pegasus output changes across versions vs CamemBERT masked-LM prediction errors." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Both concern token/processor special-token behavior, but one is runtime post-processor updates and the other is Whisper batch_decode skipping special tokens." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "Different domains: Qwen3 auto device-mapping CUDA assert on A800 vs tokenizer loading failure for EMBEDDIA/est-roberta." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "Documentation/example errors for SigLIP2 are not the same as Kosmos2.5's runtime index error on long OCR input." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "Different models and bug classes: Blip2 dtype mismatch vs Video-LLaVA missing temporal attention/shared weights." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Related model family, but one is a weight_only loading error and the other is a flash-attention generation bug; not mergeable as one fix." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription generation after align_special_tokens is unrelated to Kosmos2.5 long-input indexing." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading and TypeError compiling non-template nodes are unrelated code paths." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "LayoutLMv2 tokenizer crash and AutoTokenizer failing to load FastSpeech2ConformerTokenizer are different tokenizer/loading 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": 16, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:28218", + "issue:29159", + "issue:34634", + "issue:36010", + "issue:36331", + "issue:42175", + "issue:43262", + "issue:43526", + "issue:43577", + "issue:43582", + "issue:43606", + "issue:43644", + "issue:43650", + "issue:43653", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:43994", + "issue:44060", + "issue:44291", + "issue:44368", + "issue:44479", + "issue:44485", + "issue:44509", + "issue:44554", + "issue:44561", + "issue:44610", + "issue:44671", + "issue:44805", + "issue:44857", + "issue:44912", + "issue:44936", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45406", + "issue:45701", + "issue:45834", + "issue:45987", + "issue:46192", + "issue:46209" + ], + "soft_pairs": [ + "issue:43644|issue:43749", + "issue:43606|issue:45072", + "issue:44479|issue:45834", + "issue:44561|issue:45020", + "issue:34634|issue:36010", + "issue:43577|issue:43582", + "issue:43825|issue:43994", + "issue:44805|issue:44936", + "issue:44485|issue:44509", + "issue:45987|issue:46192", + "issue:43825|issue:43827", + "issue:44671|issue:45701", + "issue:28218|issue:29159", + "issue:44610|issue:45072", + "issue:43720|issue:45072", + "issue:43782|issue:45325", + "issue:43906|issue:46209", + "issue:42175|issue:45081", + "issue:43749|issue:44291", + "issue:43582|issue:44857", + "issue:45081|issue:45406", + "issue:43262|issue:43819", + "issue:34634|issue:36331", + "issue:43526|issue:43697", + "issue:43650|issue:43906", + "issue:44060|issue:44368", + "issue:43653|issue:44912", + "issue:43742|issue:45020", + "issue:43526|issue:44554" + ], + "prepared_review_unit_hash": "bcba21316dbae4e5fcea5ca8bd0ecd74e4889e4b5e35f707f4ff09bb9f8f6386", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36609, + "estimated_input_tokens": 9153, + "estimated_eval_tokens": 18562 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This is a heterogeneous cluster with several distinct bug families (tokenizer behavior, pipeline/v5 docs, remote_code regressions, model-specific loading/inference bugs, and training/runtime errors). None of the soft pairs look like true duplicates; most are only thematically similar.", + "confidence": 0.84, + "canonical_issue_reason": "No single canonical issue cleanly represents the whole cluster. If one must be chosen, issue 43825 is the most concrete representative of the small pipeline/v5 subset, but the cluster is too mixed for a strong canonical.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43825 is the clearest, most actionable bug report among the near-v5/pipeline items, but it still does not generalize to the unrelated tokenizer/model/training issues in the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Different failure modes: junk non-persistent buffers during save/load vs broken FSDP CPU RAM efficient loading. Same broad loading area, not the same bug." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "Both mention device/dtype issues during inference, but they affect different models and code paths (CPU offload vs bfloat16 inference). Not the same underlying defect." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Video-input regression for Qwen VL models is unrelated to Kosmos2.5 OCR index errors. Different subsystems and symptoms." + }, + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Both involve recent v5/remote_code breakage, but one is specifically `is_torch_fx_available` removal while the other is a broad regression report. Too general to merge as duplicates." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice_preset failure and GenerationMixin import error are unrelated model/API issues." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "One is a BLIP2 dtype propagation bug; the other is an Apple Silicon allocator warmup TypeError. Same platform theme only." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "Pipeline translation-task messaging and SigLIP2 nonsensical outputs are unrelated problems." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Generic mask/index shape crash versus trainer.evaluate() failing after train(). Different stack traces and code paths." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "GLM-5 RoPE discussion is unrelated to docs still mentioning removed pipeline tasks." + }, + { + "left": "issue:45987", + "right": "issue:46192", + "accept": false, + "reason": "StaticCache sequence-length contract bug and ignored return_tensors in text_kwargs are separate API issues." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "They are related to the v5 pipeline removals, but one is an error-message bug and the other is stale docs. Same topic, not the same bug." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT masked-LM regression and generic tokenization changes are not the same concrete defect." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": false, + "reason": "Both involve tokenizer behavior, but one is about an added token inserting extra space and the other is slow/fast tokenizer inconsistency more broadly. Not clearly the same bug." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "OmDet-Turbo processor size mismatch is unrelated to SwitchTransformers/TimmWrapperModel bfloat16 dtype mismatches." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Packed-weight unpacking failure during accelerate loading is a different issue from dtype mismatches in inference." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Qwen3VL weight_only loading error and Qwen2.5-VL RoPE scaling bug are separate model-specific regressions." + }, + { + "left": "issue:43906", + "right": "issue:46209", + "accept": false, + "reason": "A reproduction of an older issue and an FP8/B200 inference bug are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Backend packaging issue for `.[torch]` and a Mistral regex patch crash are unrelated." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "CPU RAM efficient loading breakage and unexpected `_is_hf_initialized` arg during init_empty_weights are distinct loading bugs." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "AppleSilicon allocator warmup TypeError and CUDA float16 AMP crash in LwDetrImageLoss are different platform/runtime failures." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Tokenizer regex patch crash and Gemma4Processor missing `_tokenizer` are separate loading/processor bugs." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "apply_chat_template sampling-rate default bug and DAC.from_latents mismatch are unrelated audio issues." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice_preset does not work and CustomTrainer.compute_loss signature breakage are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and RTDetrV2 output regression are different model/processor defects." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "'ADD THE DATA' is not a technical bug report, and it is not the same as the referenced reproduction issue." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Incorrect tied-weights warning in Qwen3-Next and a Qwen3.5 LoRA warning are related only by theme, not by identical code path." + }, + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "KeyError loading MobileLLM-125M and broad remote_code breakage are different failures." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and MPS attention correctness with differing head dims 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": 17, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:34567", + "issue:36010", + "issue:36331", + "issue:38175", + "issue:39692", + "issue:40990", + "issue:42490", + "issue:43257", + "issue:43262", + "issue:43329", + "issue:43450", + "issue:43452", + "issue:43454", + "issue:43531", + "issue:43550", + "issue:43577", + "issue:43606", + "issue:43638", + "issue:43643", + "issue:43645", + "issue:43696", + "issue:43720", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43854", + "issue:43927", + "issue:44387", + "issue:44479", + "issue:44568", + "issue:44589", + "issue:44610", + "issue:44617", + "issue:44756", + "issue:44811", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45042", + "issue:45084", + "issue:45356", + "issue:45459", + "issue:45538", + "issue:45600", + "issue:45759", + "issue:45803", + "issue:45859" + ], + "soft_pairs": [ + "issue:45084|issue:45538", + "issue:43329|issue:45600", + "issue:44589|issue:45042", + "issue:43746|issue:44811", + "issue:43638|issue:44936", + "issue:43452|issue:44991", + "issue:44387|issue:44617", + "issue:43550|issue:44912", + "issue:36010|issue:43645", + "issue:43819|issue:43927", + "issue:43257|issue:43329", + "issue:39692|issue:44912", + "issue:43854|issue:44977", + "issue:38175|issue:44610", + "issue:43720|issue:44610", + "issue:45459|issue:45803", + "issue:36010|issue:36331", + "issue:43454|issue:43577", + "issue:43531|issue:45042", + "issue:43262|issue:44811", + "issue:43257|issue:43696", + "issue:43329|issue:44479", + "issue:42490|issue:43257", + "issue:43643|issue:45759", + "issue:43329|issue:43450", + "issue:43606|issue:45859", + "issue:43749|issue:44756", + "issue:34567|issue:40990", + "issue:44568|issue:45356" + ], + "prepared_review_unit_hash": "66261d035057627a14bec6a78bf5ef57a6a2352f24533ae58a55fea168cd28fa", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36621, + "estimated_input_tokens": 9156, + "estimated_eval_tokens": 18568 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: the soft pairs are only superficially related by subsystem/theme, not by the same underlying bug or fix. No PRs are present.", + "confidence": 0.94, + "canonical_issue_reason": "No true canonical issue exists here; the items span many unrelated bugs (training, tokenization, multimodal/video, loading, offload, quantization, OOM), so any single issue would be a poor duplicate representative.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43749 is the strongest single anchor only because it has the most discussion/inbound references, but it still does not represent the rest of this mixed cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Unrelated bugs: template-node compilation vs tokenizer max-length constant." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Different code paths: multimodal video token counting vs stale auto_mappings references." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Different failures: missing Float8 storage vs PIL image processor torchvision requirement." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "Different features and code paths: local PEFT checkpoint loading vs Whisper batch_decode behavior." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Both are training/eval failures, but the underlying bugs are different (DeepSpeed ZeRO-3 index error vs evaluate-after-train failure)." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Different model loading problems: gguf_file handling vs tokenizer loading for a specific repo." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both are OOM reports, but for different models and mechanisms; not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "Unrelated: torch.compile/SDPA crash in Bamba vs MXFP4 load fallback in gpt-oss." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "Different regressions: GenerationMixin import path vs custom model initialization in Jupyter." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "Different model bugs: DAC latent reconstruction mismatch vs DiaConfig token-id persistence." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Different areas: Qwen3 MoE loading under accelerate/deepspeed vs multimodal video token counting." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "Different issues: SigLIP2 docs/example problems vs gpt-oss MXFP4 loading behavior." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Different model-specific generation/load failures; not the same concrete bug." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Different vision-model issues: zero probabilities vs processor/model input-size mismatch." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "Different code paths: packed-weight loading vs processor output shape." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both involve exception handling, but in different functions and with different bugs." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "Both affect trainer-related APIs, but one is an import error and the other is a compute_loss signature break." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "Different model loading bugs: lm_head weight tying vs dtype propagation to QFormer." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Unrelated: Qwen3-MoE sliding-window logic vs image processor torchvision dependency." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Both are audio/processor issues, but sampling-rate defaulting and batch_decode special-token handling are distinct." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Different problems: MoE weight conversion under accelerate/deepspeed vs plain OOM on GPT-oss-20b." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": false, + "reason": "Both are video-related, but one is undefined variables in token counting and the other is a broader v5.3.0 regression." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Completely different topics: trailing-slash local-path warning vs Qwen3 MoE loading." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both are config-loading issues, but one is missing fields with trust_remote_code and the other is text_config unwrapping for multimodal configs." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "Different video bugs: undefined helper variables vs incorrect batched processor output shape." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but they are different models and failure modes." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "Different loading/memory issues: FSDP CPU RAM efficient loading vs mmap on Strix Halo OOM." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "Unrelated: TrainerState token counter not updating vs GPT-oss perplexity issue." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but for different models and different root causes." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29127", + "issue:29159", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:38175", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43452", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43782", + "issue:43824", + "issue:43927", + "issue:43931", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44568", + "issue:44821", + "issue:44849", + "issue:44898", + "issue:44912", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45072", + "issue:45137", + "issue:45357", + "issue:45381", + "issue:45405", + "issue:45488", + "issue:45538", + "issue:45701", + "issue:45823", + "issue:45907" + ], + "soft_pairs": [ + "issue:43257|issue:43931", + "issue:29159|issue:34567", + "issue:42915|issue:43742", + "issue:44291|issue:45042", + "issue:42915|issue:44821", + "issue:44849|issue:45701", + "issue:44568|issue:45488", + "issue:44987|issue:45538", + "issue:43452|issue:43582", + "issue:43697|issue:45357", + "issue:41720|issue:43540", + "issue:43927|issue:44912", + "issue:42915|issue:45538", + "issue:44479|issue:45381", + "issue:33357|issue:34567", + "issue:43540|issue:44479", + "issue:42175|issue:43824", + "issue:43540|issue:44560", + "issue:43317|issue:45538", + "issue:44062|issue:44991", + "issue:45137|issue:45405", + "issue:44987|issue:45907", + "issue:24643|issue:29159", + "issue:28282|issue:34689", + "issue:43577|issue:45072", + "issue:43257|issue:43756", + "issue:43742|issue:44291", + "issue:28282|issue:29127", + "issue:42175|issue:44488", + "issue:42915|issue:43749", + "issue:38175|issue:44898", + "issue:43582|issue:43782", + "issue:43540|issue:45823" + ], + "prepared_review_unit_hash": "ce942500b578f62cb2daa85ac56ab686d0914f14c747e94ce25803ec9bfad0b2", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37558, + "estimated_input_tokens": 9390, + "estimated_eval_tokens": 19036 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: tokenizer regressions, model-loading errors, video-input bugs, dtype/device-map issues, and a few DeepSpeed/quantization reports. I only see one plausible duplicate pair in the Qwen VL video-input regression family.", + "confidence": 0.61, + "canonical_issue_reason": "Issue 45701 is the broadest and most representative anchor here: it explicitly frames a cross-version tokenization change, which is the closest shared theme among several tokenizer-related reports in an otherwise mixed set.", + "canonical_pr_reason": null, + "best_issue_reason": "45701 is the best single issue to represent the cluster because it is generic, active, and closest to the recurring regression theme; the other items are mostly narrower or unrelated one-off failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Both are Qwen loading problems, but one is an MoE weight-conversion issue under accelerate+deepspeed and the other is a Qwen3-VL weight-shape mismatch; different failure modes and code paths." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer inconsistency and TrainerState token-count tracking are unrelated subsystems." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Qwen3Moe FP8 loading failure vs MobileLLM key error are different model-loading bugs." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Init-empty-weights argument handling and PIL/torchvision backend requirements are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3Moe FP8 loading is not the same bug as AutoImageProcessor URL loading." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "Qwen3.5 hidden-state output bug is unrelated to the tokenization-version regression in 45701." + }, + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both concern tokenizer behavior, but they affect different tokenizers and different mechanisms (BOS/EOS insertion vs LlamaTokenizer pre-tokenizer override)." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "Loading failure for a specific model and CLIPTokenizer max-length behavior are not the same underlying bug." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "gguf_file loading breakage and Apple Silicon allocator warmup TypeError are unrelated." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "RTDetr output divergence and Qwen3.5 save_pretrained key corruption are different code paths." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Auto device-mapping CUDA assert on Qwen3 is unrelated to Qwen3OmniMoe video processing." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "Custom token ID persistence and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "FP8 Qwen3Moe loading and FSDP CPU RAM efficient loading are different bugs." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Both report the same Qwen VL video-input regression in v5.3.0, centered on incorrect video-position handling; they look like the same underlying code-path issue." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "MacOS CLIP bus error and TrainerState token counting are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Qwen3OmniMoe video ValueError and Qwen2.5/3.x VL video regression are different model families and likely different bugs." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Packaging/backend dependency inclusion and missing Qwen2_5_VL import are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Both are Qwen video issues, but they describe different models and different error types." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "Device-map/offload dequantization failure and CLIP tokenizer max-length behavior are unrelated." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "AddedToken special-arg duplication and tokenizer loading for est-roberta are different tokenizer bugs." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and PEFT version bump are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model loading regression and list_repo_files revision forwarding are unrelated." + }, + { + "left": "issue:24643", + "right": "issue:29159", + "accept": false, + "reason": "DeepSpeed training weight-shape error and tokenizer slow/fast inconsistency are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Missing PyTorch import and Llama 3.2 Vision loading regression are different issues." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both mention dtype, but one is BLIP2 loading state and the other is CI inference mismatches in different models." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MOE weight conversion and Smollm3 RoPE-layer count are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "MobileLLM key error and init_empty_weights constructor mismatch are different failures." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "AutoModel missing PyTorch and LayoutLMv3 error-message clarity are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "TensorFlow optional dependency packaging and a specific model-loading failure are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Same as the earlier Qwen3Moe FP8 vs FSDP mismatch: different bugs." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 and Perceiver image-classification resolution handling are unrelated." + }, + { + "left": "issue:43582", + "right": "issue:43782", + "accept": false, + "reason": "Apple Silicon allocator TypeError and Qwen3VL weight_only loading error are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:45823", + "accept": false, + "reason": "Qwen3OmniMoe video processing and Gemma4 PLE device mismatch are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative about the mostly heterogeneous cluster. The lone accepted pair (44479/45381) is plausibly the same Qwen VL video-input regression family, and the other soft-edge rejects do not overclaim duplication." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 19, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:24643", + "issue:29127", + "issue:34567", + "issue:36331", + "issue:41762", + "issue:43404", + "issue:43452", + "issue:43493", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43906", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44488", + "issue:44514", + "issue:44589", + "issue:44625", + "issue:44737", + "issue:44792", + "issue:44821", + "issue:44877", + "issue:44898", + "issue:44913", + "issue:44936", + "issue:44964", + "issue:45042", + "issue:45081", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45405", + "issue:45479", + "issue:45588", + "issue:45600", + "issue:45920", + "issue:45987", + "issue:46121", + "issue:46202" + ], + "soft_pairs": [ + "issue:44737|issue:45588", + "issue:43404|issue:45987", + "issue:43526|issue:45479", + "issue:43906|issue:45381", + "issue:43493|issue:43994", + "issue:34567|issue:36331", + "issue:43756|issue:45325", + "issue:44291|issue:44821", + "issue:44913|issue:46121", + "issue:44062|issue:45081", + "issue:44792|issue:44964", + "issue:45042|issue:45081", + "issue:45357|issue:45920", + "issue:43742|issue:44589", + "issue:43452|issue:44488", + "issue:43531|issue:44821", + "issue:43643|issue:45375", + "issue:45405|issue:45600", + "issue:44625|issue:44877", + "issue:24643|issue:29127", + "issue:43720|issue:44898", + "issue:43697|issue:45216", + "issue:43906|issue:46202", + "issue:41762|issue:44936", + "issue:43688|issue:43749", + "issue:44514|issue:45381" + ], + "prepared_review_unit_hash": "b54cd1fabff219749c1fd7104c4fb35a6d09ed42fcba9cafd1123ba56bf47140", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35911, + "estimated_input_tokens": 8978, + "estimated_eval_tokens": 18212 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is heterogeneous and the soft pairs mostly share only broad subsystem names, not the same underlying defect. I would not merge any of the proposed soft edges. The most representative issue in the cluster is the broad tokenizer regression in 45920.", + "confidence": 0.24, + "canonical_issue_reason": "issue:45920 is the broadest, most cross-cutting report here: it describes a tokenizer regression affecting several model families, making it a better cluster representative than the many narrow, model-specific bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45920 is the best global representative because it spans multiple affected models and points to a shared code-path regression rather than a one-off model-specific symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "XLNet device-placement work vs a flash-attention auxiliary-loss crash; unrelated code paths." + }, + { + "left": "issue:43404", + "right": "issue:45987", + "accept": false, + "reason": "Mistral3 weight tying vs StaticCache sequence-length return type; different bugs." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "BeitImageProcessorFast label reduction vs sequence-classification loss degeneracy; not the same defect." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "An isolated reproduction of another issue vs a Qwen2.5-VL video position-id bug; too different to treat as duplicates." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": false, + "reason": "Both concern SigLIP2, but one is HF-vs-JAX parity and the other is nonsensical pipeline output; not clearly the same code-path bug." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "TrainerState token counting vs a custom compute_loss signature error; unrelated trainer issues." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Smollm3 RoPE layer dropping vs Qwen2.5-VL temporal position scaling; distinct model-specific problems." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized mismatch vs AutoImageProcessor URL loading; different loading paths." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "GPTNeoXConfig reload resetting rotary_pct vs rope-validation TypeError on JSON-loaded lists; different serialization/validation bugs." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "AddedToken kwarg duplication vs Mistral regex backend_tokenizer AttributeError; unrelated tokenizer failures." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "janus generation test failure vs Phi-4 multimodal model loading failure; different models and symptoms." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "PIL backend torchvision dependency issue vs Mistral tokenizer regex patch crash; no shared bug." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Qwen3.5 save_pretrained visual-key regression vs tokenizer wrong token IDs across several models; different subsystems." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "MobileLLM key error vs missing Float8 storage lookup; distinct loading failures." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "gguf_file/from_pretrained handling vs failure to load cjvt/sleng-bert; not the same code path." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior vs AutoImageProcessor URL loading; unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Generic trust_remote_code config-field loss vs Qwen3_5MoeVisionConfig strict field dropping; different config issues." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "Dependency-version bump vs stale auto_mappings references; unrelated maintenance problems." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "num_labels propagation bug vs granite_speech config loading strictness; different config concerns." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Auxiliary-loss normalization vs FSDP CPU RAM efficient loading; unrelated model/training bugs." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Batched chat-template crash vs video vision_position_ids bug; distinct Qwen2.5-VL paths." + }, + { + "left": "issue:43906", + "right": "issue:46202", + "accept": false, + "reason": "Reproduction of an external issue vs Gemma pipeline roundtrip failure on a token boundary; not the same underlying bug." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading error vs trainer.evaluate() failing after train(); different lifecycle stages and failures." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "BitNet packed-weight loading bug vs Perceiver interpolation failure; unrelated." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output regression vs Qwen3.5 save_pretrained regression; different models and code paths." + }, + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "Reproduction issue vs Qwen2.5-VL video position-id scaling bug; not the same defect." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 20, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:28282", + "issue:29159", + "issue:30064", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:35141", + "issue:38175", + "issue:41720", + "issue:43295", + "issue:43452", + "issue:43531", + "issue:43577", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43742", + "issue:43749", + "issue:43819", + "issue:43906", + "issue:43927", + "issue:44361", + "issue:44442", + "issue:44464", + "issue:44466", + "issue:44484", + "issue:44493", + "issue:44514", + "issue:44568", + "issue:44704", + "issue:44743", + "issue:44811", + "issue:44843", + "issue:44936", + "issue:44991", + "issue:45005", + "issue:45290", + "issue:45356", + "issue:45372", + "issue:45440", + "issue:45542", + "issue:45701", + "issue:45704", + "issue:45823", + "issue:46050", + "issue:46170", + "issue:46178" + ], + "soft_pairs": [ + "issue:45372|issue:46170", + "issue:44466|issue:45005", + "issue:43452|issue:43577", + "issue:44493|issue:46050", + "issue:43819|issue:44811", + "issue:43452|issue:44843", + "issue:30064|issue:33666", + "issue:44936|issue:45823", + "issue:45542|issue:46178", + "issue:44484|issue:44704", + "issue:28282|issue:32090", + "issue:43295|issue:45356", + "issue:43653|issue:44361", + "issue:44361|issue:44442", + "issue:28282|issue:33357", + "issue:44514|issue:45290", + "issue:38175|issue:43295", + "issue:44464|issue:44743", + "issue:28282|issue:35141", + "issue:41720|issue:43906", + "issue:43653|issue:44568", + "issue:43577|issue:43697", + "issue:43688|issue:45704", + "issue:44843|issue:44991", + "issue:43646|issue:43749", + "issue:43688|issue:45440", + "issue:29159|issue:33357", + "issue:43531|issue:43742", + "issue:43688|issue:45701", + "issue:43653|issue:43927" + ], + "prepared_review_unit_hash": "ab4bd63918f77fb5ecb320405a4a7250cdcfbb2de872fda6065cbda30c5cb6a1", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36841, + "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 noisy and mostly heterogeneous: the soft pairs span unrelated tokenizer, training, model-loading, and device/runtime bugs. I would not merge any of the proposed soft edges.", + "confidence": 0.68, + "canonical_issue_reason": "issue:45701 is the broadest tokenizer-regression report here and has the strongest signal as an umbrella issue for the tokenizer-related subset of this cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45701 is the best single issue to anchor the cluster because it is open, broad, and likely to subsume several tokenizer-version regression reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45372", + "right": "issue:46170", + "accept": false, + "reason": "Different dependency failures: Gemma processor loading vs Windows/ROCm torch.distributed import/type issues." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights in v5, but one is a specific lm_head serialization/device bug and the other is a broader translation-model report." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Unrelated problems: gguf_file loading breakage vs BLIP2 dtype propagation." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Position-id key mismatches are not the same as sinq quantization save/load breakage." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "Different model paths and symptoms: DAC latent reconstruction vs Whisper batch decode special-token handling." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Tokenizer/model loading failures, but one is gguf_file-related and the other is an offline mistral-regex cache lookup bug." + }, + { + "left": "issue:30064", + "right": "issue:33666", + "accept": false, + "reason": "Different domains: image processor segmentation-map handling vs Qwen2-VL multi-GPU training." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer lifecycle bug vs Gemma4 device-map forward mismatch; not the same code path." + }, + { + "left": "issue:45542", + "right": "issue:46178", + "accept": false, + "reason": "Both concern backend detection, but one is TensorFlow/tensorboard setup and the other is tensorflow_text mapping." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "save_pretrained shard-size behavior is unrelated to AutoProcessor forwarding kwargs to cached_file." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for missing PyTorch is unrelated to Trainer NoneType broadcast errors." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Both are regressions, but one is processor/tokenizer API breakage and the other is Kimi codec/regex handling." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird special-token registration bug is unrelated to MLukeTokenizer AttributeError failures." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer implementations and different failure modes; no shared concrete bug." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "PyTorch import absence and MacOS bus errors on CLIP are unrelated runtime issues." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both involve apply_chat_template, but the crash conditions differ: batched padding vs tool-call assistant messages." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Siglip2 zero probabilities and processor.tokenizer regression are unrelated issues." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Chunked generation with compiled forward is not the same as recurrent state reset in Qwen3-MoE cache handling." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Missing PyTorch import and community-CLIP bus error are different failures with different triggers." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Slow/fast tokenizer inconsistency is not the same as the MacOS CLIP bus error." + }, + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to MobileLLM loading key errors." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer special-token issues, but they affect different models and different mechanics." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "BLIP2 dtype casting and RTDetrV2 output divergence are different bugs." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "Auxiliary-loss normalization and apex RMSNorm memory leak are unrelated." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but the root causes and affected models differ." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Custom model initialization breakage is not the same as FSDP CPU RAM efficient loading." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Different model families and different failure types: auxiliary-loss math vs DeepSeekV3MoE implementation divergence." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "Auxiliary-loss normalization is unrelated to the broad tokenization regression report." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "Special-token registration and DiaConfig save/load custom token ID loss 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": 21, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:28218", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:34567", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43452", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43643", + "issue:43688", + "issue:43720", + "issue:43742", + "issue:43756", + "issue:43825", + "issue:43927", + "issue:43994", + "issue:44361", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44554", + "issue:44561", + "issue:44610", + "issue:44625", + "issue:44821", + "issue:44849", + "issue:44913", + "issue:44977", + "issue:45491", + "issue:45588", + "issue:45823", + "issue:46002", + "issue:46050", + "issue:46192", + "issue:46209" + ], + "soft_pairs": [ + "issue:45491|issue:46002", + "issue:43550|issue:43927", + "issue:44464|issue:44977", + "issue:43643|issue:44561", + "issue:43550|issue:44361", + "issue:44479|issue:44625", + "issue:41720|issue:44464", + "issue:44977|issue:46209", + "issue:43577|issue:45588", + "issue:32090|issue:34567", + "issue:43577|issue:44554", + "issue:43504|issue:43550", + "issue:43504|issue:43720", + "issue:43504|issue:44610", + "issue:44466|issue:46050", + "issue:32090|issue:33666", + "issue:33666|issue:36010", + "issue:43742|issue:44821", + "issue:45823|issue:46050", + "issue:43720|issue:43927", + "issue:43606|issue:43927", + "issue:43452|issue:44451", + "issue:43262|issue:43295", + "issue:44849|issue:46209", + "issue:43278|issue:43688", + "issue:43994|issue:46192", + "issue:43525|issue:43572", + "issue:33666|issue:36331", + "issue:43720|issue:44361", + "issue:43756|issue:44913", + "issue:44625|issue:46002", + "issue:43493|issue:43825", + "issue:28218|issue:33357" + ], + "prepared_review_unit_hash": "3fcc6c5a82c56003876caf027180d5ae02392e777fdfc632582c7b20a15799e7", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 37049, + "estimated_input_tokens": 9263, + "estimated_eval_tokens": 18782 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster looks like a broad false-positive mix of unrelated issues: tokenizer spacing, trainer errors, model-loading regressions, dtype/device mismatches, and task-specific model bugs. None of the soft edges looks like the same underlying bug or change.", + "confidence": 0.93, + "canonical_issue_reason": "Issue 43504 is the cleanest concrete loading-regression report to use as an anchor, though this cluster is not a real duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "43504 is the most self-contained and actionable issue here; it has a specific failure mode and is easier to triage than the more model-specific or symptom-only reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Both mention NaNs, but one is a Gemma3 sliding-window GPU bug and the other is a classification-head checkout/init issue; different code paths." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "One is a Bamba torch.compile/SDPA failure; the other is a DiaConfig save/load field-loss bug. No shared underlying defect." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Both are generation regressions, but they involve different models and different causes (compiled forward vs flash-attention)." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Both touch trust_remote_code, but one is missing fields from AutoConfig and the other is a helper removal breaking remote-code models; not the same bug." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "A compile/SDPA model bug versus a tokenizer AttributeError on tasks; unrelated." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Both are Qwen-related, but one is a video-input regression and the other is num_labels propagation." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Qwen3 auto-device-map cudaErrorAssert and compiled-forward chunked generation inconsistency are different failures." + }, + { + "left": "issue:44977", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 flash-attention generation issues versus FP8 inference on B200; different code paths." + }, + { + "left": "issue:43577", + "right": "issue:45588", + "accept": false, + "reason": "BLIP2 dtype-loading mismatch and flash_attention.py s_aux=None handling are unrelated." + }, + { + "left": "issue:32090", + "right": "issue:34567", + "accept": false, + "reason": "Trainer broadcast TypeError versus TrainerState token-counter not updating; different trainer bugs." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "MPS attention correctness and BLIP2 dtype behavior are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Legacy-field model loading failure is not the same as the Bamba compile/SDPA regression." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Both are loading regressions, but one is a legacy-field preset load issue and the other is packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Model preset loading failure and OmDet processor/model input-size mismatch are different bugs." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Tied-weight serialization depending on device and quantized save/load breakage are not the same code-path problem." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer NoneType broadcast and Qwen2-VL multi-GPU training are unrelated." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Multi-GPU training versus GenerationMixin import failure; different subsystems." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "MobileLLM load key error and AutoImageProcessor-from-URL failure are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Gemma4 device mismatch with device_map='auto' is not the same as sinq quantized save/load breakage." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Packed-weight accelerate loading bug versus DiaConfig missing fields; unrelated." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU-offload device mismatch and config field loss are different bugs." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "gguf_file/from_pretrained breakage and ScandiBERT load failure are too different to treat as one bug." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio apply_chat_template sampling-rate default issue versus processor.tokenizer/image passthrough regression; different APIs." + }, + { + "left": "issue:44849", + "right": "issue:46209", + "accept": false, + "reason": "Qwen3.5 hidden-states bug and FP8 inference failure are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "BF16-to-FP32 dtype drift and auxiliary-loss normalization are different layers of the stack." + }, + { + "left": "issue:43994", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 output quality issue versus return_tensors being ignored with text_kwargs; not the same defect." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Both are missing-pad-token config regressions, but they affect different model configs and field names; no shared concrete bug is shown." + }, + { + "left": "issue:33666", + "right": "issue:36331", + "accept": false, + "reason": "Qwen2-VL multi-GPU training and CustomTrainer compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "BitNet packed-weight loading and MLukeTokenizer task AttributeError have no common code path." + }, + { + "left": "issue:43756", + "right": "issue:44913", + "accept": false, + "reason": "Smollm3 RoPE layering bug and GPTNeoX rotary_pct reload bug are both config-adjacent, but not the same underlying issue." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "num_labels propagation and NaNs in classification heads are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and pipeline translation-task messaging are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer extra-space after added token and MacOS bus error on a CLIP model 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": 22, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:33357", + "issue:33666", + "issue:39692", + "issue:43257", + "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:43701", + "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:44568", + "issue:44589", + "issue:44734", + "issue:44792", + "issue:44843", + "issue:44849", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45357", + "issue:45405", + "issue:45520", + "issue:45584", + "issue:45759", + "issue:45812", + "issue:45823" + ], + "soft_pairs": [ + "issue:44964|issue:45405", + "issue:44849|issue:45357", + "issue:39692|issue:43493", + "issue:44568|issue:45812", + "issue:43540|issue:44977", + "issue:43334|issue:43572", + "issue:44488|issue:44843", + "issue:43493|issue:43844", + "issue:43278|issue:43844", + "issue:45200|issue:45229", + "issue:43756|issue:45083", + "issue:39692|issue:43295", + "issue:43334|issue:45759", + "issue:43257|issue:43299", + "issue:43646|issue:44466", + "issue:43824|issue:44589", + "issue:33357|issue:33666", + "issue:43504|issue:43606", + "issue:45229|issue:45823", + "issue:43927|issue:44361", + "issue:44589|issue:45520", + "issue:43572|issue:44291", + "issue:44464|issue:45584", + "issue:43278|issue:43701", + "issue:43606|issue:44112", + "issue:43504|issue:43653", + "issue:43575|issue:45405", + "issue:44792|issue:45405", + "issue:43994|issue:44493", + "issue:44423|issue:44734" + ], + "prepared_review_unit_hash": "1380460df57be53af670159d95aa560322f38a4050a5e5e6ff266196cb904194", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36911, + "estimated_input_tokens": 9228, + "estimated_eval_tokens": 18712 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: many separate model-specific regressions and config/tokenizer bugs, with only loose topical similarity around Transformers 5.x compatibility and multimodal loading. I would not merge any of the soft pairs as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": "issue:45812 is the broadest and most user-impacting issue here (tokenization regression across Granite models) and is the closest thing to a cluster anchor, though the surrounding issues are not true duplicates of it.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45812 is the strongest global representative because it describes a concrete, high-signal regression with broad model impact and multiple references, unlike the mostly isolated model-specific bugs elsewhere in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Unrelated: model loading failure for Phi-4 multimodal vs an unreleased PEFT version bump." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Different bugs: Qwen3.5 hidden-states behavior vs incorrect save_pretrained visual encoder keys." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "One is a broken documentation/example flow; the other is a SigLIP2 implementation mismatch vs JAX." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both tokenizer-related, but different regressions and models: mDeBERTa special-token insertion vs Granite token IDs." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Different failure paths: Qwen3OmniMoe video-processing ValueError vs Qwen3.5 flash-attention generation issue." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Superficially similar missing config fields, but different models, attributes, and code paths." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "Different subsystems: model loading for sleng-bert vs offline tokenizer patching via model_info()." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 fidelity mismatch and DeepSpeed gradient inflation are unrelated bugs." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype drift in eval vs gradient growth under ZeRO-3 are distinct training issues." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Defaulting mm_token_type_ids to zeros is unrelated to Gemma4 multi-GPU OOM." + }, + { + "left": "issue:43756", + "right": "issue:45083", + "accept": false, + "reason": "Different models and symptoms: RoPE layer count bug vs a helper-length calculation issue." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Doc/example mistakes for SigLIP2 are not the same as the processor.tokenizer regression." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Both involve multimodal config handling, but one is missing pad_token_id and the other is unwrap-text_config behavior." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Related family, but different concrete problems: loading/weight conversion under accelerate+deepspeed vs Qwen3VL Moe loading failure." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage and device-dependent lm_head serialization are different regressions." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "Importing a missing model class and a Float8 storage lookup failure are unrelated." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "CLIP bus error on macOS and Qwen2-VL multi-GPU training are completely different issues." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "Different models and code paths: BEiT legacy-field loading vs bark-small CPU offload device mismatch." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but one is OOM during inference and the other is a device-placement mismatch during forward." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "Both tokenizer/config related, but the underlying failures and models differ." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage type resolution and Python 3.13 flash_attn import metadata are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "Both are Transformers 5 compatibility problems, but one is a config field rename and the other is an unexpected constructor argument." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Different generation bugs: compiled-forward chunking inconsistency vs empty-transcription handling in Whisper." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype drift and resume_from_checkpoint key mismatch do not share the same root cause." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "A real runtime device mismatch vs a stale CI test are not the same underlying bug." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "BEiT preset loading and BigBirdTokenizer special-token registration are unrelated." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "TP OOM for a large Qwen model is unrelated to the PEFT version bump issue." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "A failed Janus generation test and an unreleased dependency version bump are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 producing bad outputs and widespread unexpected position-id keys are distinct regressions." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serve/runtime crashes, but one is a multimodal .to() type error and the other is KV-cache indexing; different code paths." + }, + { + "left": "issue:44843", + "right": "issue:45357", + "accept": false, + "reason": "Offline tokenizer patching and incorrect visual-encoder save keys 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": 23, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:29942", + "issue:30333", + "issue:32090", + "issue:34567", + "issue:37428", + "issue:38175", + "issue:39692", + "issue:42490", + "issue:43232", + "issue:43295", + "issue:43450", + "issue:43493", + "issue:43540", + "issue:43572", + "issue:43582", + "issue:43696", + "issue:43701", + "issue:43824", + "issue:43844", + "issue:43873", + "issue:43957", + "issue:44186", + "issue:44521", + "issue:44554", + "issue:44610", + "issue:44849", + "issue:44964", + "issue:44977", + "issue:44987", + "issue:45072", + "issue:45084", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45356", + "issue:45381", + "issue:45406", + "issue:45446", + "issue:45520", + "issue:45588", + "issue:45823", + "issue:46050", + "issue:46121", + "issue:46192", + "issue:46209" + ], + "soft_pairs": [ + "issue:43873|issue:46050", + "issue:29942|issue:30333", + "issue:45406|issue:46121", + "issue:45588|issue:46050", + "issue:42490|issue:43696", + "issue:43844|issue:44964", + "issue:28218|issue:34567", + "issue:29159|issue:32090", + "issue:44521|issue:45290", + "issue:43582|issue:44977", + "issue:44554|issue:45446", + "issue:45084|issue:45520", + "issue:15354|issue:38175", + "issue:43540|issue:45200", + "issue:45072|issue:45198", + "issue:37428|issue:43957", + "issue:45200|issue:45823", + "issue:43232|issue:44849", + "issue:44186|issue:46192", + "issue:43493|issue:43701", + "issue:45381|issue:46209", + "issue:45823|issue:46192", + "issue:43295|issue:43450", + "issue:43295|issue:44610", + "issue:43572|issue:44987", + "issue:44554|issue:45356", + "issue:43572|issue:43957", + "issue:37428|issue:43824", + "issue:39692|issue:46192" + ], + "prepared_review_unit_hash": "89e2a9f82e740b883b1aeb94fb5cbfc79e77a143ae1a914e384ce7cc22901f06", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36422, + "estimated_input_tokens": 9106, + "estimated_eval_tokens": 18468 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous; the soft edges are mostly title-level similarities across unrelated bugs, so none of the candidate pairs look like true duplicates or mergeable PR-equivalents.", + "confidence": 0.97, + "canonical_issue_reason": "issue:29159 is the broadest, most representative issue in the set (generic tokenizer inconsistency), but the cluster is too mixed for a strong canonical duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:29159 is the best cluster anchor because it is general and closest to the tokenizer-related subset, even though the overall cluster does not form a single bug family.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Both mention quantization, but one is about offloading behavior and the other about persistence after save/load; different bugs." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "Flash Attention test failures and MLflow reporting on failed jobs are unrelated subsystems and failure modes." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer is unrelated to rope-parameter validation TypeError." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "One is a flash-attention integration AttributeError; the other is quantized-model save/load breakage." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "Local-path import warnings and CUDA OOM on GPT-oss-20b are clearly different problems." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient growth and Phi-4 multimodal load failure are not the same code-path bug." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing after added tokens is unrelated to TrainerState token-count tracking." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Slow/fast tokenizer inconsistency is unrelated to the Trainer _gpu_broadcast_one NoneType error." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Zero assistant masks in multimodal templates and crashes on tool-call messages are different chat-template bugs." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon allocator TypeError and Qwen3.5 flash-attention generation issues do not share a concrete bug." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "MPS attention correctness and a PyTorch version check/import bug are unrelated." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "Template-node compilation failure and Python 3.13 flash_attn KeyError are different import/compile paths." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "torch.jit.script GeneratorExp export failure and unexpected zero probabilities in SigLIP2 are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Qwen3OmniMoe video-processing ValueError and Gemma4 mm_token_type_ids behavior are separate model-specific issues." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "dtype mismatch in inference and Wav2Vec2 save/tokenization failures are not the same bug." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "Flash-attention utility import error and meta-device model loading breakage are distinct loading issues." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Gemma4 defaulting mm_token_type_ids and PLE device mismatch under device_map=auto are different failures." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "Generation kwargs after sync_gpus and Qwen3.5 output_hidden_states behavior are separate generation bugs." + }, + { + "left": "issue:44186", + "right": "issue:46192", + "accept": false, + "reason": "LayoutLMv2Tokenizer crashes on NER/padding and return_tensors being ignored are unrelated tokenizer API issues." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "SigLIP2 implementation discrepancy and resume_from_checkpoint key mismatch are not the same code path." + }, + { + "left": "issue:45381", + "right": "issue:46209", + "accept": false, + "reason": "Qwen2.5-VL video position IDs and FP8 inference on B200 are different model/hardware problems." + }, + { + "left": "issue:45823", + "right": "issue:46192", + "accept": false, + "reason": "Gemma4 device placement and return_tensors handling are unrelated regressions." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "Gemma4Processor _tokenizer regression and batched video shape handling are different processor issues." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "Same as above: processor tokenizer access vs OmDet-Turbo image-size mismatch are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "StableLmConfig missing pad_token_idx and transformers>=5.1.0 loading fast are different versioning/loading failures." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "MPS attention correctness bug and Kimi-K2.5 tokenizer regression are unrelated." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "Missing pad_token_idx and meta-device loading failures do not describe the same defect." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "A missing flash-attention import symbol and a missing Qwen model class import are distinct import errors." + }, + { + "left": "issue:39692", + "right": "issue:46192", + "accept": false, + "reason": "SigLIP2 documentation/example errors are unrelated to return_tensors being ignored when text_kwargs is passed." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 24, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:38175", + "issue:41720", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43334", + "issue:43388", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43827", + "issue:43844", + "issue:43856", + "issue:43873", + "issue:43901", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44442", + "issue:44466", + "issue:44514", + "issue:44521", + "issue:44610", + "issue:44792", + "issue:44857", + "issue:44898", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45325", + "issue:45412", + "issue:45520", + "issue:45563", + "issue:45584", + "issue:46170", + "issue:46178", + "issue:46192", + "issue:46209" + ], + "soft_pairs": [ + "issue:45200|issue:45325", + "issue:45520|issue:46178", + "issue:43295|issue:43873", + "issue:43550|issue:43844", + "issue:44898|issue:45412", + "issue:44466|issue:45127", + "issue:43334|issue:43525", + "issue:44792|issue:45563", + "issue:43824|issue:43975", + "issue:43450|issue:43575", + "issue:41720|issue:46209", + "issue:42915|issue:43824", + "issue:43388|issue:43526", + "issue:42490|issue:43493", + "issue:43278|issue:43856", + "issue:28218|issue:35141", + "issue:43653|issue:44112", + "issue:38175|issue:43493", + "issue:43526|issue:46192", + "issue:43975|issue:45520", + "issue:44442|issue:45198", + "issue:38175|issue:45412", + "issue:45520|issue:46170", + "issue:44514|issue:44521", + "issue:43262|issue:43450", + "issue:44206|issue:45584", + "issue:43827|issue:43901", + "issue:43819|issue:44112", + "issue:44610|issue:44857" + ], + "prepared_review_unit_hash": "93acfbd16f6cb607b323d2e8bfe4260f7147f40c15fbd1022929cdf37a13f6d0", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36622, + "estimated_input_tokens": 9156, + "estimated_eval_tokens": 18568 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous; none of the soft pairs look like true duplicates, so I rejected all of them. The only weakly central issue is the import/backend failure around `flash_attn` (45520), but even that is just the best representative of a noisy cluster, not a real duplicate hub.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 45520 is the most connected item in the set and is a concrete, reproducible bug report; it is the closest thing to a cluster anchor, though the surrounding issues are still unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "45520 is the clearest and most central issue to use as a representative for this noisy cluster, but it should not be treated as a true duplicate canonical without stronger evidence.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Different bugs in different multimodal paths: Gemma 4 text-only token-type defaults vs Qwen2.5-VL temporal position scaling." + }, + { + "left": "issue:45520", + "right": "issue:46178", + "accept": false, + "reason": "Both are import/backend-related, but they hit different backend symbols and different failure modes; not the same underlying bug." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "One is a processor API regression, the other is quantization/offloading behavior; too far apart." + }, + { + "left": "issue:43550", + "right": "issue:43844", + "accept": false, + "reason": "Different code paths and symptoms: torch.compile with SDPA vs ZeRO-3 gradient blow-up." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Unrelated issues: Perceiver resize/interpolation failure vs RT-DETR memory not being freed." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both mention tied weights, but one is device-dependent serialization of `lm_head.weight` and the other is LoRA merge collapse with extended vocab; not the same bug." + }, + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Same broad symptom (`pad_token_id` missing), but for different model configs and likely different fixes." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "Different failures: a Janus image-generation test vs a stale paged-generate warning." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "Import error for a missing model class vs detokenization regression in DeepSeek Coder; no shared code-path evidence." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Video processor batching shape bug vs tensor-parallel OOM on a large model; unrelated." + }, + { + "left": "issue:41720", + "right": "issue:46209", + "accept": false, + "reason": "Both are hardware-specific inference failures, but on different models and different accelerator constraints; not mergeable as one fix." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "Qwen3MoE FP8 failure and Qwen2.5-VL import error are unrelated." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both concern labels, but one is metric gathering dropping tuple elements and the other is BeitImageProcessorFast reducing labels incorrectly; different behaviors." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "Path-trailing-slash import warning is unrelated to SigLIP2 implementation discrepancies." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "Training/eval dtype mismatch vs MoE memory inefficiency; different defects." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing after added tokens and embedding reinitialization after resize are different issues." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "BigBirdTokenizer special-token registration bug vs GraniteSpeech CI test staleness; not the same bug." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "Both mention SigLIP2, but one is a specific zero-probability symptom and the other is a broader HF-vs-JAX discrepancy; insufficient evidence of one underlying bug." + }, + { + "left": "issue:43526", + "right": "issue:46192", + "accept": false, + "reason": "Label reduction bug and `return_tensors` being ignored are unrelated processor issues." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Detokenization regression in a model tokenizer is unrelated to the `flash_attn` import_utils KeyError." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "AutoTokenizer loading failure for one tokenizer and save_pretrained/tokenization failure for another are separate tokenizer bugs." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 zero-probabilities bug vs RT-DETR memory cleanup bug; unrelated." + }, + { + "left": "issue:45520", + "right": "issue:46170", + "accept": false, + "reason": "Different dependency/import problems: `flash_attn` backend mapping vs torch.distributed on Windows/ROCm." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both touch `apply_chat_template`, but one crashes on batched padding=False input and the other returns zero assistant masks; different concrete failures." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio processor sample-rate defaulting and batched video processor shapes are unrelated." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "Feature extractor `center` arg crash and Whisper empty-transcription generation failure are different code paths." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs regressions around pipeline deprecations, but they refer to different pages and API details; too broad to merge as one fix." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "DAC latent/forward mismatch is unrelated to a GraniteSpeech CI test issue." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Different model families and failures: OmDet-Turbo input-size mismatch vs LwDetr AMP/CUDA crash." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 25, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:32090", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:39692", + "issue:42490", + "issue:43278", + "issue:43335", + "issue:43425", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43720", + "issue:43819", + "issue:43825", + "issue:43854", + "issue:43931", + "issue:44112", + "issue:44206", + "issue:44351", + "issue:44360", + "issue:44448", + "issue:44493", + "issue:44561", + "issue:44743", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:45072", + "issue:45083", + "issue:45127", + "issue:45200", + "issue:45245", + "issue:45290", + "issue:45310", + "issue:45381", + "issue:45478", + "issue:45588", + "issue:45593", + "issue:45823", + "issue:45859", + "issue:45920", + "issue:46192" + ], + "soft_pairs": [ + "issue:43819|issue:45083", + "issue:37428|issue:43425", + "issue:43335|issue:43550", + "issue:43606|issue:45083", + "issue:44857|issue:45593", + "issue:29942|issue:32090", + "issue:44448|issue:44743", + "issue:45127|issue:45245", + "issue:37428|issue:44351", + "issue:45290|issue:45381", + "issue:45200|issue:45588", + "issue:43278|issue:43825", + "issue:43335|issue:43653", + "issue:43611|issue:43931", + "issue:44206|issue:45823", + "issue:43550|issue:44112", + "issue:43819|issue:44493", + "issue:44855|issue:45083", + "issue:43335|issue:43504", + "issue:45588|issue:45859", + "issue:29127|issue:32090", + "issue:43335|issue:45083", + "issue:39692|issue:43526", + "issue:39692|issue:44493", + "issue:45310|issue:45478", + "issue:43720|issue:44112", + "issue:43335|issue:44112", + "issue:43335|issue:45072", + "issue:44448|issue:44779", + "issue:44360|issue:44561", + "issue:42490|issue:43854", + "issue:45920|issue:46192", + "issue:43606|issue:44857", + "issue:34689|issue:35141" + ], + "prepared_review_unit_hash": "3ed5a871e59065f61a6c0c8dbbb4af2b1be0a50e6163bfc081a226e5b14d244d", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 38095, + "estimated_input_tokens": 9524, + "estimated_eval_tokens": 19304 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These items are mostly unrelated issues with only one clear duplicate pair: the Qwen3.5 Moe from_pretrained regression (#45310 and #45478). All other soft edges look like different bugs or different code paths and should be rejected.", + "confidence": 0.94, + "canonical_issue_reason": "issue:45310 is the earlier, fuller report of the only clear duplicate pair and cleanly captures the Qwen3.5 Moe from_pretrained regression.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the best issue anchor because it is the strongest true duplicate target in the set; the rest are heterogeneous and do not form a coherent duplicate cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "Different model subsystems and failure modes: DAC latents/STE mismatch vs qwen3_omni_moe feature length helper behavior." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "One is an import error for a missing symbol; the other is a broad torch version compatibility complaint." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "SwitchTransformers sparse-layer config bug vs Bamba torch.compile/SDPA failure; unrelated code paths." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU offload device mismatch in bark-small vs a qwen3_omni_moe helper-function behavior issue." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "Float16 AMP crash in LwDetrImageLoss is unrelated to D-FINE missing auxiliary losses." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures vs Trainer/_gpu_broadcast_one NoneType error are different bugs." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Pegasus output regression across versions vs recurrent state reset in qwen3 modular code; different mechanisms." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA/extended-vocab collapse is unrelated to the 2^24 category limit error." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import-related, but they involve different missing symbols and different breakages." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "Chat template tool-call crash vs qwen2.5-vl video vision_position_ids mismatch; not the same bug." + }, + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 mm_token_type_ids defaulting issue is unrelated to flash_attention s_aux=None crashes." + }, + { + "left": "issue:43278", + "right": "issue:43825", + "accept": false, + "reason": "Embedding dtype drift during eval vs pipeline translation-task warning; different API surfaces." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Sparse layer misconfiguration in SwitchTransformers vs BigBirdTokenizer special-token registration problem." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Base model prefix loading regression is unrelated to Qwen3-VL weight-shape mismatch." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "LaserFeatureExtractor center-arg regression vs Gemma4 PLE device mismatch; different failures." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Bamba torch.compile SDPA bug vs GraniteSpeech stale device-override CI failure are unrelated." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "DAC forward/from_latents mismatch is not the same as unexpected position-id key warnings." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 IndentationError in DebertaV2 import is unrelated to qwen3_omni_moe output-length helper behavior." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "SwitchTransformers sparse-layer creation bug vs Beit semantic-segmentation preset loading issue; different models and causes." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Both touch flash/offload behavior, but one is an AttributeError on s_aux=None and the other is CPU offload incompatibility in rotary embedding." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMv3 error-message clarity issue vs trainer NoneType broadcast crash; different underlying bugs." + }, + { + "left": "issue:43335", + "right": "issue:45072", + "accept": false, + "reason": "SwitchTransformers sparse-layer bug vs dtype mismatches in bfloat16 inference; not the same defect." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "SigLIP2 docs/example errors are unrelated to BeitImageProcessorFast reduce_labels returning the wrong labels." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "Documentation/example failures vs position-id key regression; different artifacts and code paths." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Same title, same model family, same from_pretrained error; clear duplicate reports of the same regression." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "BitNet packed-weight unpacking failure is unrelated to GraniteSpeech CI device-override staleness." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "Different areas entirely: SwitchTransformers config vs GraniteSpeech CI." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Pegasus v4/v5 output change is not the same as DeepSeek tokenizer tokenization regression." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA indexer lacking ReLU vs removal of is_torch_fx_available; unrelated breakages." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Local-model path import warning vs failing to load GLM-4.7-Flash in unit tests; different issues." + }, + { + "left": "issue:45920", + "right": "issue:46192", + "accept": false, + "reason": "AutoTokenizer wrong token IDs across models vs return_tensors being ignored when text_kwargs is passed; similar API area but different bugs." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU offload device mismatch in Bark-small vs float16 AMP crash in LwDetrImageLoss; not the same code path." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Model loading regression for Llama 3.2 Vision is unrelated to embedding reinitialization after resizing." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet and stays conservative on soft-edge pairs. The only clearly supported duplicate pair is issue:45310 with issue:45478, which share the same title and model-specific from_pretrained regression. The rest of the pair rejections are consistent with the titles/bodies and do not overstate the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 26, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:28218", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:40444", + "issue:41720", + "issue:42490", + "issue:43278", + "issue:43329", + "issue:43381", + "issue:43412", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43701", + "issue:43723", + "issue:43746", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:44448", + "issue:44462", + "issue:44493", + "issue:44554", + "issue:44561", + "issue:44568", + "issue:44779", + "issue:44857", + "issue:44977", + "issue:44987", + "issue:45072", + "issue:45083", + "issue:45198", + "issue:45229", + "issue:45372", + "issue:45406", + "issue:45440", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45803", + "issue:46209" + ], + "soft_pairs": [ + "issue:38175|issue:44493", + "issue:44977|issue:45229", + "issue:44857|issue:45072", + "issue:43746|issue:46209", + "issue:40444|issue:43526", + "issue:44448|issue:44977", + "issue:28218|issue:32090", + "issue:43701|issue:45083", + "issue:32090|issue:33357", + "issue:29942|issue:35141", + "issue:43550|issue:45083", + "issue:43723|issue:44554", + "issue:44561|issue:44779", + "issue:37428|issue:44987", + "issue:43927|issue:45698", + "issue:43873|issue:45229", + "issue:44462|issue:44561", + "issue:43278|issue:43329", + "issue:42490|issue:43575", + "issue:10000|issue:15354", + "issue:41720|issue:45229", + "issue:44977|issue:45563", + "issue:44568|issue:45491", + "issue:45372|issue:45588", + "issue:44857|issue:45588", + "issue:43412|issue:45588", + "issue:45198|issue:45803", + "issue:45372|issue:45406", + "issue:43278|issue:43381", + "issue:43976|issue:45440" + ], + "prepared_review_unit_hash": "4548fdb31031631ecb74d141f908fc81a76b4989bbbf10f1900867dadf49b5d4", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36797, + "estimated_input_tokens": 9200, + "estimated_eval_tokens": 18656 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: the soft-linked issues span unrelated tokenizer, model-loading, attention, quantization, export, and runtime errors. None of the soft pairs look like the same underlying bug or change.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical issue: the linked issues cover different codepaths and symptoms, so there is no single representative duplicate target.", + "canonical_pr_reason": null, + "best_issue_reason": "No global best issue: nothing here clearly serves as an umbrella for the cluster, because the soft edges do not indicate a shared root cause.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Different problems: one is model-specific zero probabilities, the other is a position-id key warning across many models." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Flash-attention generation failure and multi-GPU CUDA OOM are distinct runtime issues." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "Separate dtype/backend regressions in different models and codepaths; not the same bug." + }, + { + "left": "issue:43746", + "right": "issue:46209", + "accept": false, + "reason": "Local checkpoint loading with PEFT adapters is unrelated to FP8 inference failing on B200." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "Multimodal iterable-dataset batching failure is not the same as BEiT processor label reduction." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Pegasus v4/v5 output drift and Qwen flash-attention generation failure are different defects." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer spacing after added tokens has nothing to do with Trainer GPU broadcast NoneType errors." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to the qwen3_omni_moe helper-function bug." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Trainer broadcast NoneType and MacOS bus error on CLIP are different failure modes." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "FlashAttention test failures and embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is not the same as undefined helper names in qwen3_omni_moe." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "AutoTokenizer v5 loading behavior and an MPS attention correctness issue are different problems." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Removal of an internal utility breaking trust_remote_code is unrelated to DeepSeek tokenizer output regression." + }, + { + "left": "issue:37428", + "right": "issue:44987", + "accept": false, + "reason": "Both are loading/import regressions, but they point to different concrete failures and codepaths." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Save/load state loss of custom token IDs is not the same as loading the wrong custom module after save_pretrained." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Quantization offloading behavior and a multi-GPU OOM are different memory-management issues." + }, + { + "left": "issue:44462", + "right": "issue:44561", + "accept": false, + "reason": "Tokenizer.json handling and a missing utility function are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype drift between train/eval is not the same as undefined video-token helper names." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "Import warning for trailing-slash paths and tensor-parallel OOM are unrelated." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "Different historical bugs: German DistilBert issue vs JIT GeneratorExp export failure." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Auto device-mapping CUDA assert and multi-GPU OOM are distinct failures." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Flash-attention generation malfunction and a stale warning for num_return_sequences are not the same issue." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Special-token insertion bug and Gemma3 NaNs under sliding-window attention are different defects." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Gemma4 processor import failure is unrelated to a flash_attention AttributeError on s_aux=None." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "Loss/AMP crash and flash_attention sink-less-model crash are different codepaths." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Executorch export failures are unrelated to the flash_attention s_aux=None attribute error." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Wav2Vec2 save/tokenization failure and a bare-except swallowing errors in Fuyu are distinct bugs." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Both touch Gemma4, but one is an import dependency issue and the other is a missing _tokenizer attribute in serve." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Python version compatibility and DeepSeekV3MoE output divergence 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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:25251", + "issue:29127", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43065", + "issue:43066", + "issue:43257", + "issue:43278", + "issue:43381", + "issue:43421", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44265", + "issue:44361", + "issue:44442", + "issue:44448", + "issue:44462", + "issue:44466", + "issue:44479", + "issue:44743", + "issue:44871", + "issue:44977", + "issue:45198", + "issue:45245", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45712", + "issue:45823", + "issue:46121" + ], + "soft_pairs": [ + "issue:45588|issue:45823", + "issue:44448|issue:44466", + "issue:25251|issue:32090", + "issue:45459|issue:45712", + "issue:43653|issue:44743", + "issue:45245|issue:45563", + "issue:44206|issue:44479", + "issue:43065|issue:43994", + "issue:45372|issue:45459", + "issue:44186|issue:45198", + "issue:43065|issue:45412", + "issue:43611|issue:46121", + "issue:44871|issue:44977", + "issue:32090|issue:33290", + "issue:44361|issue:45198", + "issue:42490|issue:43873", + "issue:33357|issue:39692", + "issue:29942|issue:30064", + "issue:29127|issue:34567", + "issue:42915|issue:42994", + "issue:43994|issue:45412", + "issue:44265|issue:45198", + "issue:43066|issue:44462", + "issue:43257|issue:43278", + "issue:43756|issue:45440", + "issue:43421|issue:45491", + "issue:43381|issue:43701", + "issue:43653|issue:44442" + ], + "prepared_review_unit_hash": "f79b8f677af473d96df026ffa27938c17ca92c5437efe37e9878e0142f459d1b", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36178, + "estimated_input_tokens": 9045, + "estimated_eval_tokens": 18346 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated Transformers bug reports; the only loose common thread is tokenizer/loading regressions, and none of the soft pairs look like true duplicates. I\u2019d use 44462 as the closest canonical issue because it is the broadest user-facing loading regression in that subset.", + "confidence": 0.32, + "canonical_issue_reason": "44462 is the most broadly scoped, user-facing tokenizer-loading regression here, so it best represents the small cluster of tokenizer-related reports better than the more niche failures.", + "canonical_pr_reason": null, + "best_issue_reason": "44462 is the best representative issue overall because it captures a common loading-path failure with wider duplicate potential than the more model-specific or one-off bugs in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Different code paths and symptoms: flash-attention handling of s_aux=None versus a Gemma4 device_map/device-placement mismatch." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Both are v5 regressions, but one is Pegasus generation output drift and the other is serialization of lm_head.weight; not the same concrete bug." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline top_k nesting and Trainer _gpu_broadcast_one NoneType errors are unrelated failures." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Tokenizer exception masking when protobuf is missing is unrelated to cleanup of leftover dummy classes." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "BigBirdTokenizer special-token decode bug and Qwen recurrent-state reset logic are different subsystems and failure modes." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "Quantized-save category limit and a stale generate() warning for num_return_sequences are not the same bug." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Unsupported center arg in a feature extractor is distinct from the Qwen video-input regression." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "Sam3PixelDecoder dummy Conv2d and SigLIP2 nonsensical outputs are different concrete model issues." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Gemma4 processor import failure due to ReasoningEffort is unrelated to protobuf-related tokenizer error masking." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "Both involve tokenizers, but LayoutLMv2 NER/padding crashes and Wav2Vec2 save_pretrained/tokenization failures are different bugs." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "A dummy layer in Sam3PixelDecoder is not the same as RT-DETR memory not being released after deletion." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "base_model_prefix loading regression and JSON-loaded rope validation list TypeError affect different code paths." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Gemma-3 eos_token_id config mismatch and Qwen3.5 flash-attention generation failure are unrelated." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "NoneType broadcast failure in Trainer is not the same as an OOM with Adafactor in DeepSpeed." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "MLukeTokenizer AttributeError and Wav2Vec2 save/tokenization failure are different tokenizer bugs." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Local-path trailing-slash import warning and quantization offloading failure are not the same issue." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "A MacOS bus error on a CLIP model and SigLIP2 documentation example errors are different problems." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention 2 test failures and void segmentation-map processing failures do not share the same root cause." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "LayoutLMv3 error clarity and TrainerState num_input_tokens_seen not updating are unrelated." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "FineGrainedFP8Config failure and quantized-model saving failure are distinct quantization bugs." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "SigLIP2 wrong outputs and RT-DETR memory leaks are different symptoms and code paths." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export.export failure with torch_compilable_check is unrelated to Wav2Vec2 tokenization/save_pretrained failures." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but one is a decoder-type mismatch while the other is AutoTokenizer ignoring tokenizer.json; not the same concrete bug." + }, + { + "left": "issue:43257", + "right": "issue:43278", + "accept": false, + "reason": "Missing MOE weight conversion under accelerate+deepspeed and BF16-to-FP32 dtype drift are different issues." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Smollm3 dropping RoPE layers and DeepseekV3MoE diverging from remote implementation are architecture-specific and not the same bug." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Runtime post-processor updates for special tokens and Gemma3 NaN embeddings with mixed-length batching are unrelated." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Eval-mode gradient checkpointing restriction and resume_from_checkpoint key mismatch are separate training-state problems." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "Both are tokenizer-related, but one is special-token registration/empty decode and the other is AutoTokenizer class loading; different root causes." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:33357", + "issue:35141", + "issue:36010", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43334", + "issue:43381", + "issue:43452", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43701", + "issue:43761", + "issue:43824", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44186", + "issue:44206", + "issue:44279", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44610", + "issue:44625", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45103", + "issue:45245", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45491", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45834", + "issue:45859", + "issue:45907" + ], + "soft_pairs": [ + "issue:44610|issue:45412", + "issue:43824|issue:44279", + "issue:29127|issue:33357", + "issue:45440|issue:45593", + "issue:29942|issue:36010", + "issue:43761|issue:45859", + "issue:43526|issue:45834", + "issue:43526|issue:44186", + "issue:45103|issue:45372", + "issue:29127|issue:35141", + "issue:44625|issue:45584", + "issue:44442|issue:45072", + "issue:43334|issue:43531", + "issue:42915|issue:43931", + "issue:43575|issue:43701", + "issue:43526|issue:44857", + "issue:42490|issue:43381", + "issue:44977|issue:45245", + "issue:43575|issue:44387", + "issue:43262|issue:44206", + "issue:44464|issue:45491", + "issue:43646|issue:44077", + "issue:44743|issue:44811", + "issue:43452|issue:45588", + "issue:44062|issue:45907", + "issue:43761|issue:44079" + ], + "prepared_review_unit_hash": "6629c151882e98d7a8fb5e428079005988cc3918d8f3bb7b52be1c8a01d22c2c", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 26, + "serialized_chars": 34908, + "estimated_input_tokens": 8727, + "estimated_eval_tokens": 17710 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft pairs are mostly unrelated issues across different models and subsystems; none look like the same underlying bug or change, so no pair should be merged as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "No single issue is a clear duplicate anchor for this set: the reports span unrelated regressions and components. If a representative must be chosen, issue 43646 is the broadest, but it is not a true canonical duplicate target.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43646 is the most general, high-level regression in the set (a broad Transformers 5.0.0 initialization break), so it is the best representative bucket if one is needed, though the cluster is not actually homogeneous.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44610", + "right": "issue:45412", + "accept": false, + "reason": "Different failures: processor input size mismatch for OmDet-Turbo vs RT-DETR memory not being released." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "One is a specific missing import for Qwen2.5-VL; the other is a broad dependency issue with no concrete same bug." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error messaging clarity and a MacOS bus error on CLIP are unrelated." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepSeekV3MoE implementation divergence and D-FINE auxiliary-loss behavior are different code paths." + }, + { + "left": "issue:29942", + "right": "issue:36010", + "accept": false, + "reason": "Flash Attention 2 test failures are not the same issue as a GenerationMixin import error." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIPVision hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are separate bugs." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BEiT FastImageProcessor label reduction and Kosmos2.5 long OCR index errors are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "These affect different tokenizer/processor pipelines and different failure modes." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "auto_docstring annotation handling and Gemma 4 loading/import issues are not the same bug." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "LayoutLMv3 box validation messaging and token-embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and Whisper empty-transcription generation failure are different problems." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "FastSpeech2Conformer tokenizer loading and bfloat16 dtype mismatches are distinct issues." + }, + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id for Qwen3-VL loading is not the same as Qwen3-MoE sliding-window behavior." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "FineGrainedFP8Config failure for Qwen3Moe and Qwen3-VL weight-shape mismatch are different model/config bugs." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "Tensor-parallel OOM on Qwen2-57B is unrelated to resume_from_checkpoint key mismatch." + }, + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "Label reduction bug in BeitImageProcessorFast and float16 AMP crash in LwDetrImageLoss are separate." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Trailing-slash local model loading warning is unrelated to gradient checkpointing being disallowed in eval mode." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Flash-attention generation problems and the 2^24 category limit are unrelated." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both mention OOM, but one is tensor-parallel model loading and the other is int4 quantization memory growth; different root causes." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio chat-template sampling-rate defaults and an unsupported center argument in LasrFeatureExtractor are unrelated." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency and Gemma3 NaN embeddings from sliding-window padding are different bugs." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Broad custom model initialization regression and a patchtsmixer post_init policy issue do not share the same concrete failure." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Qwen3.5 recurrent-state caching bug and Whisper batch_decode skip_special_tokens handling are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:45588", + "accept": false, + "reason": "gguf_file loading behavior and flash_attention sink-less AttributeError are separate code paths." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken keyword duplication and list_repo_files revision forwarding are unrelated API bugs." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVision hidden_states regression and ModelOutput key assignment with None are different issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 29, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:33290", + "issue:34567", + "issue:34689", + "issue:35141", + "issue:36010", + "issue:36331", + "issue:42490", + "issue:42907", + "issue:43232", + "issue:43316", + "issue:43381", + "issue:43388", + "issue:43493", + "issue:43525", + "issue:43611", + "issue:43643", + "issue:43650", + "issue:43720", + "issue:43723", + "issue:43749", + "issue:43761", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44077", + "issue:44188", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44743", + "issue:44855", + "issue:44871", + "issue:44898", + "issue:44913", + "issue:44987", + "issue:45003", + "issue:45137", + "issue:45216", + "issue:45290", + "issue:45440", + "issue:45479", + "issue:45698", + "issue:45759", + "issue:45803", + "issue:46178", + "issue:46202", + "issue:46209" + ], + "soft_pairs": [ + "issue:33290|issue:34689", + "issue:44442|issue:44898", + "issue:45003|issue:45440", + "issue:33290|issue:34567", + "issue:33290|issue:35141", + "issue:43723|issue:45479", + "issue:43232|issue:43388", + "issue:43381|issue:43873", + "issue:42490|issue:44743", + "issue:43316|issue:43525", + "issue:43761|issue:44188", + "issue:34689|issue:43611", + "issue:44479|issue:45290", + "issue:43720|issue:44442", + "issue:44987|issue:46178", + "issue:44077|issue:44479", + "issue:44442|issue:45803", + "issue:43493|issue:43827", + "issue:43931|issue:45216", + "issue:34689|issue:36010", + "issue:42907|issue:45698", + "issue:33290|issue:36331", + "issue:43316|issue:45759", + "issue:45290|issue:46209", + "issue:43873|issue:44871", + "issue:43749|issue:43931", + "issue:43643|issue:44913", + "issue:43650|issue:46202", + "issue:44190|issue:44496", + "issue:44855|issue:45137" + ], + "prepared_review_unit_hash": "7385df6bf9cca8b91f177942b6cdbce531036e9f2138c63cc06b7bcdc290b26c", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36890, + "estimated_input_tokens": 9223, + "estimated_eval_tokens": 18702 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is highly heterogeneous and the soft edges are mostly superficial title-level similarities rather than true duplicates. I would reject every proposed merge pair.", + "confidence": 0.83, + "canonical_issue_reason": "No single canonical issue stands out: these reports span unrelated bugs across loading, configs, training, generation, vision, quantization, and docs, so there is no clear duplicate anchor for the whole set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is the strongest standalone anchor among the set because it is currently open, detailed, and has the most discussion/inbound references, but it still is not related to most of the other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Different bugs: DeepSpeed/Adafactor OOM vs Llama 3.2 Vision model loading regression." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Unrelated failures: tokenizer loading vs Perceiver image classification resize/interpolation bug." + }, + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "Different code paths: unsafe sys.modules access vs DeepSeekV3MoE behavioral divergence." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "OOM in optimizer/deepspeed is unrelated to TrainerState token counter not updating." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Optimizer memory blowup is unrelated to embedding resize/post_init reinitialization." + }, + { + "left": "issue:43723", + "right": "issue:45479", + "accept": false, + "reason": "Tokenizer loading regression and degenerate classification loss are different problems." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation sync/update bug is unrelated to metric gathering label truncation." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Gradient checkpointing eval-mode restriction is unrelated to quantization/offloading behavior." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Local-path import warning and recurrent-state cache reset are unrelated." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "Both mention config API issues, but different models and different missing fields." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIP hidden_states regression is unrelated to attention-kernel divergence under torch.compile." + }, + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both are model-loading regressions, but they concern different mechanisms and different failure modes." + }, + { + "left": "issue:44479", + "right": "issue:45290", + "accept": false, + "reason": "Video-input regression and chat-template tool-call crash are separate bugs." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "BitNet accelerate loading issue is unrelated to tokenizer loading failure." + }, + { + "left": "issue:44987", + "right": "issue:46178", + "accept": false, + "reason": "Different symptoms and subsystems: model loading regression vs missing BACKENDS_MAPPING entry." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "Optional post_init contract issue is unrelated to video input regression." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "Tokenizer load failure and bare-except error swallowing in Fuyu are unrelated." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "Model implementation discrepancy and docs still referencing pipeline() are not the same bug." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Shape-mismatch load error and save_pretrained checkpoint corruption are distinct issues." + }, + { + "left": "issue:34689", + "right": "issue:36010", + "accept": false, + "reason": "Model loading breakage and GenerationMixin import error are different failures." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Dequantized save failure and wrong custom module reloading are separate save/load problems." + }, + { + "left": "issue:33290", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed OOM and CustomTrainer compute_loss signature mismatch are unrelated training bugs." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Config API discrepancy and AutoModelForCausalLM unwrapping behavior are different issues." + }, + { + "left": "issue:45290", + "right": "issue:46209", + "accept": false, + "reason": "Chat template crash is unrelated to FP8 inference failure on B200." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading bug and Gemma eos_token_id inconsistency are unrelated." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "FSDP CPU RAM efficient loading regression is not the same as Qwen3-VL shape mismatch." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "trust_remote_code missing-fields issue and rotary_pct reload regression are different config serialization bugs." + }, + { + "left": "issue:43650", + "right": "issue:46202", + "accept": false, + "reason": "Placeholder/noise issue is unrelated to Gemma pipeline roundtrip failure." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Local dataset loading problem is unrelated to unrecognized custom model config." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 import parsing error and DeepSpeed ZeRO3 deque pop 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": 30, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:25251", + "issue:28282", + "issue:32090", + "issue:35141", + "issue:41950", + "issue:42617", + "issue:43065", + "issue:43295", + "issue:43316", + "issue:43388", + "issue:43404", + "issue:43441", + "issue:43452", + "issue:43475", + "issue:43540", + "issue:43550", + "issue:43572", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43673", + "issue:43688", + "issue:43701", + "issue:43716", + "issue:43742", + "issue:43761", + "issue:43825", + "issue:44077", + "issue:44162", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44534", + "issue:44617", + "issue:44743", + "issue:44792", + "issue:44871", + "issue:44877", + "issue:44913", + "issue:45127", + "issue:45200", + "issue:45584", + "issue:45803", + "issue:46050", + "issue:46170", + "issue:46178" + ], + "soft_pairs": [ + "issue:25251|issue:28282", + "issue:43295|issue:43825", + "issue:43065|issue:44617", + "issue:43550|issue:43701", + "issue:43606|issue:43701", + "issue:43316|issue:44877", + "issue:43316|issue:43716", + "issue:42617|issue:43575", + "issue:43540|issue:44514", + "issue:44077|issue:44534", + "issue:44162|issue:44792", + "issue:16998|issue:22355", + "issue:44871|issue:45584", + "issue:43575|issue:43742", + "issue:43404|issue:43441", + "issue:45803|issue:46170", + "issue:43295|issue:44871", + "issue:43673|issue:44743", + "issue:41950|issue:43475", + "issue:43761|issue:44479", + "issue:43388|issue:43688", + "issue:43452|issue:44496", + "issue:43646|issue:44913", + "issue:44871|issue:46050", + "issue:45127|issue:45200", + "issue:43572|issue:46178", + "issue:32090|issue:35141" + ], + "prepared_review_unit_hash": "9cb918df43bbb13c11e4be14696883a850a55d8a56b194d35689caadb0a03657", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35722, + "estimated_input_tokens": 8931, + "estimated_eval_tokens": 18118 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of unrelated bug reports across many subsystems; the soft-similarity pairs do not look like true duplicates or mergeable same-fix candidates. No PRs are present.", + "confidence": 0.96, + "canonical_issue_reason": "issue:43646 is the most general and high-level regression in the set (custom model initialization breaking under Transformers 5.0), so it is the closest thing to a cluster anchor, though the cluster is not a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best representative issue because it describes a broad, user-facing regression with clear actionability and relatively generic impact compared with the other highly specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Different failures: pipeline output shape change vs missing `transformers.onnx` module import." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "One is a processor API regression; the other is an incorrect pipeline error message about translation support." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "`Sam3PixelDecoder` dummy layer issue is unrelated to SAM3 video CUDA OOM." + }, + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba is unrelated to checkpoint resume key mismatches." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch does not match resume-from-checkpoint key mismatch." + }, + { + "left": "issue:43316", + "right": "issue:44877", + "accept": false, + "reason": "Gemma3 config API discrepancy is unrelated to strict config loading for granite_speech." + }, + { + "left": "issue:43316", + "right": "issue:43716", + "accept": false, + "reason": "Config API mismatch is not the same as an image preprocessor dtype mismatch." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "Failed script execution and TP OOM are different problems and code paths." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Qwen3OmniMoe video-processing error is separate from batched chat-template padding crash." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "Optional `post_init` policy issue is unrelated to non-persistent buffer initialization bugs." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 regression and Janus image-generation test failure do not describe the same underlying bug." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Model max length question is unrelated to missing ONNX submodule import." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "Gemma EOS config inconsistency is unrelated to Whisper empty-transcription generation failure." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "Tensor-parallel OOM and MobileLLM config key error are different loading failures." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Untied `lm_head` weights and FlashAttention failure in Ministral-3 are different code-path problems." + }, + { + "left": "issue:45803", + "right": "issue:46170", + "accept": false, + "reason": "Bare `except` in Fuyu tensor conversion is unrelated to Windows/ROCm distributed import issues." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Processor/tokenizer regression is not the same as Gemma EOS token configuration mismatch." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Cache missing in generation and recurrent state reset are related thematically but not the same concrete bug." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Video-classification processor selection bug is unrelated to missing `fpn_position_embeddings` on SAM3 output." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "CLIP vision hidden-states regression is separate from Qwen video-input regression." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Metric label truncation and auxiliary-loss normalization are different training bugs." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "GGUF loading breakage is different from missing `model_type` in config loading." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Custom model init regression is broader and different from GPTNeoX rotary_pct persistence on reload." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "Gemma EOS mismatch is unrelated to sinq quantization save/load breakage." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA merge collapse with extended vocab is unrelated to Gemma 4 multimodal token type defaults." + }, + { + "left": "issue:43572", + "right": "issue:46178", + "accept": false, + "reason": "StableLm missing config field and tensorflow_text backend mapping error are unrelated compatibility issues." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error is unrelated to embedding reinitialization after resizing." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:35141", + "issue:41720", + "issue:42490", + "issue:43065", + "issue:43335", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43540", + "issue:43550", + "issue:43644", + "issue:43761", + "issue:43827", + "issue:43931", + "issue:43976", + "issue:43994", + "issue:44060", + "issue:44077", + "issue:44079", + "issue:44190", + "issue:44242", + "issue:44464", + "issue:44479", + "issue:44521", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44737", + "issue:44792", + "issue:44855", + "issue:44913", + "issue:44933", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45245", + "issue:45357", + "issue:45362", + "issue:45381", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45698", + "issue:45803", + "issue:45987" + ], + "soft_pairs": [ + "issue:43504|issue:45698", + "issue:44734|issue:45406", + "issue:44617|issue:45584", + "issue:45563|issue:45584", + "issue:43525|issue:43531", + "issue:43761|issue:44242", + "issue:44792|issue:44977", + "issue:43931|issue:44060", + "issue:44479|issue:44734", + "issue:44534|issue:44913", + "issue:44190|issue:44964", + "issue:44792|issue:45600", + "issue:44464|issue:45563", + "issue:43827|issue:43994", + "issue:44855|issue:45588", + "issue:44521|issue:45381", + "issue:35141|issue:42490", + "issue:41720|issue:45245", + "issue:43065|issue:44610", + "issue:45803|issue:45987", + "issue:44933|issue:45362", + "issue:43335|issue:44855", + "issue:43644|issue:44913", + "issue:44077|issue:44079", + "issue:43976|issue:44855", + "issue:43550|issue:44190", + "issue:43540|issue:44734", + "issue:45127|issue:45357", + "issue:44737|issue:45803" + ], + "prepared_review_unit_hash": "eebe10e693dc9a620a877971965034125b32849eb8455322da1ec319f8de7c01", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36404, + "estimated_input_tokens": 9101, + "estimated_eval_tokens": 18458 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster looks heterogeneous: the soft edges are mostly topical lookalikes, not true duplicates. I\u2019d reject every pair and treat issue:44521 as the best single representative only because it has the strongest activity signal, not because it matches the others.", + "confidence": 0.84, + "canonical_issue_reason": "issue:44521 has the strongest discussion/inbound-reference signal in this set, so it is the best representative if a single issue must be chosen, even though the cluster is not a real duplicate group.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the most active and well-documented item here, making it the most suitable standalone anchor among otherwise unrelated reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Both concern model loading/save behavior, but they describe different failures and code paths." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "Both involve transformers serve crashes, but one is KV-cache continuation indexing and the other is a processor attribute error." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Unrelated bugs: video-model OOM versus Whisper empty transcription after token alignment." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Different issues in generation: a stale warning versus a real failure on empty transcription." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Different configs and symptoms: missing pad_token_id versus sliding-window handling in Qwen3-MoE." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "One is a CLIPVisionModel regression; the other is an MoE load-balancing-loss bug." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "A janus image-generation test failure is not the same bug as Qwen3.5 flash-attention misgeneration." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Different model-loading failures: weight-shape mismatch versus incorrect tied-weights warning." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Both touch generation/runtime behavior, but one is a video-input regression and the other is KV-cache continuation indexing." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent-buffer initialization and GPTNeoX rotary_pct reload are different root causes." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading in a trainer script is unrelated to loading a multimodal model preset." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "A failing janus generation test and stale auto_mappings references are different problems." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency is not the same as a stale num_return_sequences warning." + }, + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Docs still referencing removed pipeline APIs is unrelated to a broken SigLIP model output." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python parsing/import failure and flash_attention s_aux handling are separate defects." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "Both involve multimodal chat/video behavior, but the bugs are different: assistant mask generation versus vision_position_ids." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Token-embedding reinit after resize and a trailing-slash import warning are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3 auto device-mapping CUDA asserts are unrelated to the category-limit RuntimeError." + }, + { + "left": "issue:43065", + "right": "issue:44610", + "accept": false, + "reason": "A dummy Conv2d in Sam3PixelDecoder and a processor/model input-size mismatch are different bugs." + }, + { + "left": "issue:45803", + "right": "issue:45987", + "accept": false, + "reason": "Bare except in tensor conversion and StaticCache.get_seq_length() returning the wrong type are unrelated." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Missing image_utils import and Qwen3.5 chat crashes are different code paths and symptoms." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "A sparse-layer config bug and a Python 3.13 import/parsing failure are not the same issue." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "The non-persistent-buffer junk regression is distinct from GPTNeoX rotary_pct being lost on reload." + }, + { + "left": "issue:44077", + "right": "issue:44079", + "accept": false, + "reason": "Optional post_init handling and ModelOutput key assignment are unrelated framework bugs." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "A Python-version compatibility issue is not the same as a DebertaV2Model indentation/parsing regression." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure is unrelated to dataset loading in a training script." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Qwen3OmniMoe video-input ValueError and serve KV-cache indexing crash are different failures." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "LoRA merge collapse with extended vocab and incorrect visual-encoder save keys are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "XLNet CPU arange device placement and a bare-except swallowing errors 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": 32, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36246", + "issue:38175", + "issue:38617", + "issue:39692", + "issue:41628", + "issue:41950", + "issue:42175", + "issue:42907", + "issue:43065", + "issue:43295", + "issue:43335", + "issue:43404", + "issue:43441", + "issue:43479", + "issue:43531", + "issue:43643", + "issue:43650", + "issue:43854", + "issue:43976", + "issue:44016", + "issue:44112", + "issue:44186", + "issue:44190", + "issue:44220", + "issue:44246", + "issue:44279", + "issue:44462", + "issue:44496", + "issue:44561", + "issue:44683", + "issue:44734", + "issue:44938", + "issue:44964", + "issue:44987", + "issue:45003", + "issue:45216", + "issue:45245", + "issue:45276", + "issue:45335", + "issue:45375", + "issue:45446", + "issue:45542", + "issue:45584", + "issue:45600", + "issue:45632", + "issue:45907", + "issue:46082", + "issue:46121" + ], + "soft_pairs": [ + "issue:44112|issue:45600", + "issue:43335|issue:44190", + "issue:44462|issue:45907", + "issue:46082|issue:46121", + "issue:39692|issue:44190", + "issue:44938|issue:45542", + "issue:45276|issue:45584", + "issue:43295|issue:45276", + "issue:43479|issue:45375", + "issue:36246|issue:41628", + "issue:45245|issue:45584", + "issue:43650|issue:44220", + "issue:44186|issue:44734", + "issue:42175|issue:44279", + "issue:36246|issue:38617", + "issue:41950|issue:44016", + "issue:44246|issue:45003", + "issue:44683|issue:45446", + "issue:43404|issue:45216", + "issue:38175|issue:43065", + "issue:44016|issue:44220", + "issue:43643|issue:44496", + "issue:44964|issue:45600", + "issue:44112|issue:44964", + "issue:43854|issue:44964", + "issue:44734|issue:45632", + "issue:44561|issue:45632", + "issue:43976|issue:44987", + "issue:43531|issue:45600", + "issue:42907|issue:43441", + "issue:43295|issue:45335" + ], + "prepared_review_unit_hash": "a4bd2e9513250d84aaab85b622c21f62e9116bcc531e94792a7bc2c27f755572", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36960, + "estimated_input_tokens": 9240, + "estimated_eval_tokens": 18736 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: it mixes import/API breakages, model-specific regressions, CI/doc issues, and unrelated config/runtime bugs. No soft pair looks like the same concrete fixable bug or change.", + "confidence": 0.84, + "canonical_issue_reason": "Issue 41628 is the clearest broad, user-facing API regression in the set: a top-level import failure that fits the recurring import-breakage theme better than the narrower model/script-specific reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 41628 is the best representative because it describes a concrete, high-impact Transformers import regression with a clear repro and wider applicability than the more specialized issues around it.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "Both are CI-related, but one is a GraniteSpeech test failure and the other is a stale auto_mappings reference; different bugs and different fixes." + }, + { + "left": "issue:43335", + "right": "issue:44190", + "accept": false, + "reason": "A SwitchTransformers config bug and an image-classification training script dataset-loading problem are unrelated." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "One is AutoTokenizer ignoring tokenizer.json, the other is list_repo_files not forwarding revision; both touch repo/file loading but not the same bug." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "Llama head_dim validation and rope-parameter JSON list handling are distinct config-validation failures." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "A SigLIP2 docs example error and a local dataset loading failure in a training script are not the same issue." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 load failure and a missing TensorFlow backend when only tensorboard is installed are unrelated environment/backend problems." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Gemma4 embedding resize propagation and Whisper empty-transcription generation are different model behaviors." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Processor/tokenizer regression and Gemma4 token-embedding resize semantics are separate API bugs." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Default multimodal config initialization and missing deepstack_visual_indexes serialization are different config issues." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they are for different symbols/modules and look like separate export problems." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "Category cardinality limits and Whisper generation on empty output are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "A placeholder issue and an fbank feature extraction bug are not related." + }, + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "LayoutLMv2 tokenizer crashes and serve/KV-cache tensor indexing failures affect different code paths." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "Missing TensorFlow in a torch install and a vague dependency issue are not the same concrete bug." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "Both are ImportErrors, but they involve different missing symbols in different modules; the fixes would not be merged together." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "A video-classification pipeline bug and a notebook syntax error are unrelated." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Import performance complaints and unsafe sys.modules access may be loosely related, but they are not the same underlying bug." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Both are flex_attention-related, but one is a compiled-op failure on torch >= 2.9 and the other is a version-check/import gating bug." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Weight tying in Mistral3 and incorrect save_pretrained output for Qwen3.5 are different model serialization/training issues." + }, + { + "left": "issue:38175", + "right": "issue:43065", + "accept": false, + "reason": "Zero probabilities in SigLIP2 and a dummy Conv2d in Sam3PixelDecoder are unrelated." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "Notebook syntax and audio feature extraction are different failures." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Missing fields from trust_remote_code configs and unrecognized model/type handling are different config-loading problems." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "A Phi-4 multimodal load regression and a stale Sam3LiteText mapping reference are separate model/config issues." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "CI test flakiness in GraniteSpeech and a Phi-4 loading regression are different bugs." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "GLM-4.7-Flash unit-test loading failure and Phi-4 multimodal loading failure are model-specific, not the same bug." + }, + { + "left": "issue:44734", + "right": "issue:45632", + "accept": false, + "reason": "Serve/KV-cache tensor indexing and trust_remote_code cache-path collisions are unrelated." + }, + { + "left": "issue:44561", + "right": "issue:45632", + "accept": false, + "reason": "A removed helper breaking trust_remote_code models and a local-model cache collision are different trust_remote_code problems." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility and loading a specific model under transformers>=5.1.0 are not the same concrete issue." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and stale auto_mappings references are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized Ministral/Devstral models and FlashAttention failures are different issues in the same family only by name." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Processor/tokenizer regression and decoder.embed_tokens resize propagation are distinct API behaviors." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 33, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:38175", + "issue:39692", + "issue:41950", + "issue:42490", + "issue:43295", + "issue:43299", + "issue:43335", + "issue:43381", + "issue:43441", + "issue:43479", + "issue:43525", + "issue:43526", + "issue:43577", + "issue:43650", + "issue:43673", + "issue:43824", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44075", + "issue:44242", + "issue:44279", + "issue:44336", + "issue:44351", + "issue:44368", + "issue:44496", + "issue:44521", + "issue:44617", + "issue:44779", + "issue:44933", + "issue:45020", + "issue:45042", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45325", + "issue:45335", + "issue:45357", + "issue:45405", + "issue:45584", + "issue:45588", + "issue:45593", + "issue:45632", + "issue:45759", + "issue:45812", + "issue:45823", + "issue:46209" + ], + "soft_pairs": [ + "issue:42490|issue:45632", + "issue:43526|issue:43901", + "issue:43673|issue:44336", + "issue:43441|issue:45588", + "issue:44496|issue:45042", + "issue:43335|issue:45593", + "issue:44016|issue:44062", + "issue:43824|issue:44933", + "issue:44075|issue:44368", + "issue:43299|issue:45216", + "issue:43381|issue:45593", + "issue:43901|issue:44496", + "issue:43976|issue:44279", + "issue:43650|issue:46209", + "issue:44496|issue:45759", + "issue:45335|issue:45584", + "issue:43479|issue:43525", + "issue:44242|issue:45200", + "issue:44521|issue:45325", + "issue:43299|issue:45357", + "issue:44351|issue:45278", + "issue:43577|issue:45216", + "issue:38175|issue:44617", + "issue:41950|issue:45020", + "issue:39692|issue:43901", + "issue:44279|issue:45405", + "issue:44779|issue:45812", + "issue:41950|issue:45823", + "issue:43295|issue:44779" + ], + "prepared_review_unit_hash": "c133a80a7025d31b50e5fbc1e24b25a8cfdcff66b610781a95f1b441f0b9a2a8", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36389, + "estimated_input_tokens": 9098, + "estimated_eval_tokens": 18452 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is very heterogeneous: the soft pairs mostly share broad topical language (import errors, multimodal configs, tokenizer regressions, flash-attention, docs) but not the same concrete bug. I would reject all soft edges and treat issue:45278 as only a loose import-regression anchor, not a true duplicate canonical.", + "confidence": 0.9, + "canonical_issue_reason": "No single canonical duplicate set stands out. If forced to anchor the cluster, issue:45278 is the broadest import-regression report, but the rest of the items are too diverse to justify a true canonical duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best overall anchor because it is the broadest and most central of the few import-related reports; the other issues are mostly unrelated one-off bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42490", + "right": "issue:45632", + "accept": false, + "reason": "Both are path/loading-related, but one is a slash-end warning and the other is a trust_remote_code cache collision; different bugs." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "Image-label reduction bug vs pipeline docs mismatch; unrelated behavior and code paths." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Generation cache issue in chunked prefill vs ANSI-color output in loading_report; no shared bug." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Both mention FlashAttention, but one is a model loading/failure report and the other is an AttributeError on s_aux=None; too different to merge." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Model config import failure vs PIL image processor requiring torchvision; unrelated." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction bug vs D-FINE auxiliary-loss omission; different models and logic." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error vs tokenizer AddedToken keyword conflict; separate issues." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Qwen2.5-VL import failure vs nonexistent image_utils import; both are import problems but not the same module or bug." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "SGD optimizer args ignored vs tie_word_embeddings warning; unrelated." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3VL model-loading regression vs Qwen3.5 save_pretrained regression; same family, different code paths and symptoms." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode vs D-FINE auxiliary losses; no overlap beyond being model bugs." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Pipeline docs note vs unrecognized model/config loading error; unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python version compatibility problem vs generic dependency issue; not the same bug." + }, + { + "left": "issue:43650", + "right": "issue:46209", + "accept": false, + "reason": "Placeholder/data request vs FP8 inference failure; unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized model config vs AutoModelForCausalLM not unwrapping text_config; distinct multimodal-loading problems." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "Token embedding resize bug vs Whisper empty-transcription generation failure; different subsystems." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Phi4Multimodal default-config initialization vs Llama4 missing pad_token_id; both config issues but not the same defect." + }, + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "MoE load-balancing loss gating bug vs Gemma4 mm_token_type_ids defaulting; different training behaviors." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "All-zero assistant mask bug vs rope-index temporal scaling bug; both multimodal but not the same underlying change." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3.5/Qwen3VL load regression vs save_pretrained visual-key regression; related family, different failures." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "HybridCache import error may be one symptom of broader import regressions, but the titles are too broad/specific to confirm the same underlying bug." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "BLIP2 dtype propagation bug vs Qwen3.5 checkpoint save regression; unrelated." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "SigLIP2 zero-probabilities issue vs Sam3Video OOM; completely different problems." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Video-classification pipeline processor lookup bug vs remote_code breakage in recent versions; not the same issue." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "SigLIP2 docs example errors vs text-classification docs mismatch; both docs-related but different examples and failures." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "Generic dependency issue vs MIN_PEFT_VERSION bump/release problem; unrelated." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Deepseek tokenizer regression vs Granite tokenizer regression; both tokenizer bugs, but different model families and likely different fixes." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Video-classification processor lookup bug vs Gemma4 PLE device mismatch; unrelated." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Processor/tokenizer regression with multimodal inputs vs Deepseek tokenizer wrong results; different code paths and model classes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 34, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:33290", + "issue:42175", + "issue:42994", + "issue:43065", + "issue:43262", + "issue:43299", + "issue:43317", + "issue:43441", + "issue:43450", + "issue:43650", + "issue:43653", + "issue:43673", + "issue:43723", + "issue:43761", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44162", + "issue:44261", + "issue:44263", + "issue:44279", + "issue:44291", + "issue:44297", + "issue:44361", + "issue:44462", + "issue:44466", + "issue:44521", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:44898", + "issue:45081", + "issue:45276", + "issue:45397", + "issue:45405", + "issue:45542", + "issue:45593", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45823", + "issue:45834", + "issue:45907", + "issue:46132" + ], + "soft_pairs": [ + "issue:44683|issue:45542", + "issue:43299|issue:43854", + "issue:43976|issue:45405", + "issue:43650|issue:46132", + "issue:44462|issue:45081", + "issue:43450|issue:44279", + "issue:44297|issue:44462", + "issue:44016|issue:44291", + "issue:45081|issue:45907", + "issue:44466|issue:45276", + "issue:43441|issue:45834", + "issue:45600|issue:45712", + "issue:42994|issue:43317", + "issue:43653|issue:45593", + "issue:33290|issue:45656", + "issue:44623|issue:45397", + "issue:44623|issue:44749", + "issue:43065|issue:44898", + "issue:44261|issue:44263", + "issue:44361|issue:45593", + "issue:43441|issue:44683", + "issue:44016|issue:45823", + "issue:44671|issue:44779", + "issue:42175|issue:43976", + "issue:43723|issue:44162", + "issue:43901|issue:43994", + "issue:43262|issue:44521", + "issue:43673|issue:43761" + ], + "prepared_review_unit_hash": "2d226b252c2693489c5d52387b60d993a4440c3bae14104c035adb38cd540d75", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36010, + "estimated_input_tokens": 9003, + "estimated_eval_tokens": 18262 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The items do not form a true duplicate cluster; they span many unrelated bugs across tokenizers, model loading, processors, docs, and packaging. I would treat issue 44521 as the best representative issue, but none of the soft-edge pairs look like the same underlying bug/change.", + "confidence": 0.92, + "canonical_issue_reason": "Best representative issue here: it is a concrete, reproducible API regression in `apply_chat_template`, has multiple inbound references, and is broader/more actionable than the many model-specific or unrelated reports.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44521 is the strongest standalone issue: clear symptom, good downstream impact, active discussion, and a specific code-path to fix.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "Different subsystems and failure modes: compiled flex attention on new torch vs missing TensorFlow backend installation." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Both are model-loading problems, but they concern different models and likely different code paths." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Unrelated packaging/version regressions: Python support vs an over-bumped PEFT minimum version." + }, + { + "left": "issue:43650", + "right": "issue:46132", + "accept": false, + "reason": "No shared bug: one is a placeholder/data issue, the other is a behavioral change in attention registration." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "Both involve tokenizers, but one is AutoTokenizer precedence for tokenizer.json and the other is a Mistral regex-patch crash." + }, + { + "left": "issue:43450", + "right": "issue:44279", + "accept": false, + "reason": "Completely unrelated: video processor batching shape bug vs a general dependency issue." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "Different bugs in tokenizer serialization vs tokenizer loading precedence; related area, not the same underlying defect." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error and init_empty_weights TypeError are unrelated." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "Different issues: Mistral tokenizer regex patch crash vs list_repo_files revision forwarding in CI." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "Both touch Gemma4, but one is tied-weight serialization and the other is resize_token_embeddings behavior." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "Different models and symptoms: FlashAttention failure vs OCR index error." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both affect repository maintenance/CI, but they are separate file/reference problems with no shared code path." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Both touch quantization/loading, but one is saving a quantized model and the other is dequantized offload loading." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer special-token decoding bug vs D-FINE auxiliary-loss behavior; unrelated." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Both mention DeepSpeed/optimizer, but the concrete bugs differ: Adafactor OOM vs optimizer.step being called twice." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Processor save-pretrained file omission and Gemma-4 zero3 from_pretrained are different regressions." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "No match: missing processor save files vs performance regression after upgrade." + }, + { + "left": "issue:43065", + "right": "issue:44898", + "accept": false, + "reason": "Different vision bugs in different models: dummy Conv2d placeholder vs Perceiver interpolation failure." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Different model components and symptoms: missing rms_norm_eps precision drift vs torch.split return-value handling." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer attribute error vs D-FINE loss behavior; unrelated." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "Different feature failures: model-specific FlashAttention regression vs compiled flex_attention failing on torch 2.9+." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Both are model-specific wrong-results reports, but they concern different tokenizers/models and distinct regressions." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Packaging/backend dependency mismatch vs Python version support breakage; not the same bug." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "AutoTokenizer loading regression and ESM2 breakage are different issues." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Docs mismatch for a pipeline option vs incorrect model/pipeline outputs; not the same underlying change." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Same API name, but different modalities and symptoms: default 16kHz audio processing vs multimodal assistant mask generation." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Generation cache regression and CLIPVision hidden_states bug 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": 35, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36246", + "issue:41628", + "issue:41950", + "issue:42491", + "issue:42757", + "issue:42886", + "issue:43299", + "issue:43352", + "issue:43450", + "issue:43525", + "issue:43526", + "issue:43531", + "issue:43596", + "issue:43643", + "issue:43673", + "issue:43784", + "issue:43824", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:44188", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44297", + "issue:44336", + "issue:44351", + "issue:44373", + "issue:44485", + "issue:44496", + "issue:44561", + "issue:44589", + "issue:44617", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:45003", + "issue:45250", + "issue:45292", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45561", + "issue:45563", + "issue:45593", + "issue:46132" + ], + "soft_pairs": [ + "issue:43784|issue:45542", + "issue:43824|issue:43976", + "issue:43450|issue:44617", + "issue:44297|issue:45397", + "issue:43525|issue:43881", + "issue:43901|issue:44246", + "issue:43525|issue:43643", + "issue:44188|issue:44336", + "issue:41628|issue:43976", + "issue:44589|issue:45362", + "issue:44336|issue:44373", + "issue:42491|issue:43299", + "issue:45561|issue:45563", + "issue:44295|issue:45397", + "issue:44246|issue:44496", + "issue:44242|issue:45593", + "issue:43531|issue:45362", + "issue:41950|issue:44220", + "issue:43526|issue:45003", + "issue:42886|issue:46132", + "issue:36246|issue:42757", + "issue:36246|issue:44351", + "issue:44336|issue:44485", + "issue:44246|issue:44871", + "issue:45003|issue:45561", + "issue:43673|issue:44561", + "issue:44779|issue:45292", + "issue:36246|issue:43596", + "issue:44295|issue:44749", + "issue:44242|issue:44857", + "issue:43352|issue:45250" + ], + "prepared_review_unit_hash": "ea79cb538e4fdc291f985295c7e0e9840a170a2efb34e5f2bafba8e320daa6e8", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36692, + "estimated_input_tokens": 9173, + "estimated_eval_tokens": 18602 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly loose semantic similarity rather than true duplicates. I rejected all soft edges, and there are no PRs in the set. If forced to pick a representative, issue:44246 is the least-bad broad node, but it is only a weak canonical.", + "confidence": 0.66, + "canonical_issue_reason": "issue:44246 is the broadest, most generic report in the cluster and sits on multiple soft-similarity edges, so it is the best available representative despite the cluster being heterogeneous.", + "canonical_pr_reason": null, + "best_issue_reason": "Same as canonical: it is the most general issue and the closest thing to a central triage anchor here, though not a true duplicate hub.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Both are import/backend failures, but the symbols and failure modes differ (`nn` import vs TensorBoard/TensorFlow backend)." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Missing Qwen2.5-VL export vs Python version compatibility; same package, different bugs." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Video processor shape bug vs SAM video OOM; different symptoms and code paths." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save config mismatch is unrelated to Gemma-4 ZeRO3 loading." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Missing `pad_token_id` on Llama4Config is unrelated to glm-4v-9b loading failure." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Docs mismatch for `return_all_scores` is not the same as intermittent import slowness." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Llama4 config attribute error and `trust_remote_code` missing fields are different failures." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence and ANSI output in loading_report are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "AutoImageProcessor import error is not the same as the Python 3.9/3.10 support issue." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage lookup failure and Qwen3.5 chat crash are different model/runtime problems." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "ANSI code leakage and a wrong docstring do not share the same underlying bug." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both mention Qwen3-MoE, but one is hf4.x/hf5.x LoRA compatibility and the other is a v5 loading regression." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "A pytest-xdist race on `captured_info.txt` is unrelated to a stale `num_return_sequences` warning." + }, + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Buffered `position_ids` access and Gemma-4 ZeRO3 loading are different code paths." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import performance issue is unrelated to an unrecognized model/config error." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "These are both auxiliary-loss omissions, but for different models and different gating flags; not one concrete bug." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Qwen3-MoE `sliding_window` behavior and Qwen3.5 chat crashes are not the same issue." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification image-processor lookup and `_torch_extract_fbank_features()` are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "`reduce_labels` returning one label and unsafe `sys.modules` access are different bugs." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline cache loading and `AttentionInterface.register` behavior change are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different libraries." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Different missing `transformers` exports; same package, not the same bug." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI formatting leakage and GLM-5 RoPE discussion are unrelated." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import slowness and Gemma eos_token_id inconsistency are different problems." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe `sys.modules` access is unrelated to the xdist file race." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Both touch v5 remote-code breakage, but the concrete defects differ (`GenerationMixin` cache vs removed helper)." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Deepseek tokenizer correctness regression and `resize_token_embeddings` not updating output embeddings are different bugs." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Qwen2.5-VL import error and BertModel/deepspeed IndexError are unrelated." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "Buffered `position_ids` handling and slower filtering after upgrade are not the same issue." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "MoE loss gating and an AMP crash in LwDetrImageLoss are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "A model-specific Flash Attention 2 support gap is not the same as the generic Flash Attention 2 issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 36, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:42886", + "issue:43065", + "issue:43232", + "issue:43257", + "issue:43262", + "issue:43408", + "issue:43475", + "issue:43540", + "issue:43653", + "issue:43901", + "issue:43931", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44188", + "issue:44246", + "issue:44336", + "issue:44368", + "issue:44466", + "issue:44479", + "issue:44483", + "issue:44496", + "issue:44877", + "issue:44977", + "issue:44987", + "issue:44995", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45229", + "issue:45292", + "issue:45335", + "issue:45375", + "issue:45397", + "issue:45412", + "issue:45459", + "issue:45563", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45656", + "issue:45701", + "issue:45759", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46170", + "issue:46202" + ], + "soft_pairs": [ + "issue:44162|issue:45200", + "issue:43901|issue:45003", + "issue:44496|issue:45003", + "issue:45412|issue:46123", + "issue:44877|issue:45375", + "issue:43262|issue:44246", + "issue:45292|issue:45584", + "issue:43257|issue:45137", + "issue:45292|issue:45701", + "issue:45375|issue:45759", + "issue:45987|issue:46082", + "issue:43232|issue:44336", + "issue:44987|issue:45459", + "issue:44479|issue:46170", + "issue:43065|issue:43408", + "issue:44995|issue:45200", + "issue:43475|issue:43540", + "issue:44466|issue:45335", + "issue:45397|issue:45656", + "issue:42886|issue:45229", + "issue:44483|issue:46202", + "issue:44060|issue:44077", + "issue:44188|issue:46202", + "issue:44995|issue:45588", + "issue:45137|issue:45600", + "issue:43653|issue:44162", + "issue:44368|issue:44977", + "issue:44877|issue:46170", + "issue:44368|issue:45563", + "issue:43931|issue:44368" + ], + "prepared_review_unit_hash": "415d1b11b004e591ae623baad14771715efc3a5d173196de66df367e745cf206", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36748, + "estimated_input_tokens": 9187, + "estimated_eval_tokens": 18630 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is highly heterogeneous: the soft pairs mostly share only broad subsystem vocabulary (embeddings, config loading, deepspeed, video, tokenizers) rather than the same concrete bug. I reject all soft edges as non-duplicates.", + "confidence": 0.94, + "canonical_issue_reason": "If forced to choose one representative, issue 44466 is the closest hub because several nearby items revolve around tied-weight / embedding behavior. That said, the cluster does not form a single coherent duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44466 is the most concrete and broadly scoped among the recurring embedding/tied-weight reports, so it is the best single issue to anchor from this mixed set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "Different models and failures: ESM2 is broken vs Gemma4 mm_token_type_ids defaulting." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch in a pipeline vs unsafe sys.modules access in modeling_utils." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model/config loading vs sys.modules access bug; unrelated code paths." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "Memory leak on RT-DETR deletion vs final-batch pipeline flag bug." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Both involve config strictness, but different models and different missing fields." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate default vs slow import; no shared bug." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "Resize-token-embeddings/output head sync vs Whisper empty transcription after align_special_tokens." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "DeepSpeed-related, but one is weight conversion and the other is a zero3 deque crash." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Embedding resize bug vs tokenization changing across versions; too broad to merge." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Missing config field under @strict vs from_config not unwrapping text_config." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "Cache-length return type bug vs LlamaConfig head_dim validation; unrelated." + }, + { + "left": "issue:43232", + "right": "issue:44336", + "accept": false, + "reason": "Generation kwargs/sync_gpus bug vs ANSI escape output in loading_report." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Model load failure in v5.1+ vs tokenizer error masking when protobuf is absent." + }, + { + "left": "issue:44479", + "right": "issue:46170", + "accept": false, + "reason": "Video-input regression vs Windows/ROCm distributed import failure; different stacks." + }, + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "SAM3 dummy Conv2d implementation vs model-type mismatch warning; not the same defect." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "Stale cache crash in GlmMoeDsa vs Gemma4 token-type defaults." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "SAM3 vision encoder attribute error vs Qwen3OmniMoe video processing ValueError." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Both touch embeddings, but one is serialization inconsistency and the other is resize_token_embeddings propagation." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma-4 zero3 loading vs duplicate optimizer.step under deepspeed; distinct bugs." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache load failure vs Gemma4 multi-GPU CUDA OOM." + }, + { + "left": "issue:44483", + "right": "issue:46202", + "accept": false, + "reason": "Chat/completions API rejection vs Gemma pipeline roundtrip on \u2581; unrelated symptoms." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Incorrect tied-weights warning vs patchtsmixer post_init allowance; different configs." + }, + { + "left": "issue:44188", + "right": "issue:46202", + "accept": false, + "reason": "torch.compile attention-kernel divergence vs token roundtrip failure; no shared code path." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Second-forward stale cache crash vs flash_attention s_aux=None AttributeError." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed zero3 deque underflow vs stale auto_mappings reference to removed configs." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer special-token decode bug vs ESM2 model breakage." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "Tied-embeddings warning vs flash-attention generation regression; not the same failure." + }, + { + "left": "issue:44877", + "right": "issue:46170", + "accept": false, + "reason": "Config strictness issue vs Windows ROCm import/dependency issue." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "Embedding warning vs stale num_return_sequences warning; different warnings and triggers." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch vs tied_word_embeddings warning; 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": 37, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:41628", + "issue:42371", + "issue:42994", + "issue:43065", + "issue:43232", + "issue:43475", + "issue:43482", + "issue:43525", + "issue:43643", + "issue:43646", + "issue:43673", + "issue:43701", + "issue:43827", + "issue:43873", + "issue:44038", + "issue:44162", + "issue:44188", + "issue:44297", + "issue:44361", + "issue:44530", + "issue:44623", + "issue:44661", + "issue:44734", + "issue:44749", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44945", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45230", + "issue:45310", + "issue:45356", + "issue:45375", + "issue:45381", + "issue:45397", + "issue:45406", + "issue:45459", + "issue:45478", + "issue:45656", + "issue:46077", + "issue:46123", + "issue:46132", + "issue:46170", + "issue:46202" + ], + "soft_pairs": [ + "issue:43673|issue:44734", + "issue:44530|issue:46170", + "issue:43065|issue:46123", + "issue:44933|issue:45459", + "issue:44297|issue:44623", + "issue:44361|issue:44749", + "issue:44869|issue:45356", + "issue:43646|issue:44162", + "issue:43232|issue:43673", + "issue:44623|issue:45656", + "issue:44749|issue:45656", + "issue:43673|issue:43827", + "issue:43701|issue:44749", + "issue:44871|issue:46132", + "issue:45230|issue:45397", + "issue:43475|issue:45381", + "issue:42994|issue:43482", + "issue:43475|issue:43643", + "issue:28282|issue:30990", + "issue:44038|issue:44623", + "issue:46077|issue:46123", + "issue:45406|issue:46202", + "issue:42371|issue:45092", + "issue:41628|issue:44933", + "issue:43475|issue:44661", + "issue:42994|issue:43525", + "issue:45375|issue:45478", + "issue:45081|issue:45103", + "issue:45310|issue:45375", + "issue:43873|issue:44945", + "issue:43232|issue:44188" + ], + "prepared_review_unit_hash": "86466317d89bfac30aa9c0af7797f3e48649c91733898a8228a4beac43c768e3", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36879, + "estimated_input_tokens": 9220, + "estimated_eval_tokens": 18696 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: most items are unrelated bugs that only share broad transformer subdomains (loading, generation, tokenizers, serving, quantization). The soft edges look like thematic similarity, not true duplicate pairs.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the broadest, most active report in the set, but the cluster is too mixed for it to serve as a true duplicate canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Different generation failures: missing cache state vs wrong tensor indexing in serve/KV continuation." + }, + { + "left": "issue:44530", + "right": "issue:46170", + "accept": false, + "reason": "PagedAttention crash on Qwen3.5 models is unrelated to a Windows/ROCm torch.distributed import dependency issue." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "SAM3 PixelDecoder dummy Conv2d is unrelated to MaskGenerationPipeline dropping a final partial batch." + }, + { + "left": "issue:44933", + "right": "issue:45459", + "accept": false, + "reason": "Missing import from image_utils is a different bug than protobuf-related tokenizer error masking." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "Both involve save_pretrained, but tokenizer metadata mismatch and processor missing files are different code paths." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "Tokenizer AttributeError on tasks is unrelated to the Chinese performance slowdown report." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper timestamp decode trailing-character crash is a different tokenizer regression than Kimi-K2.5 codec handling." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "AutoConfig trust_remote_code missing fields is unrelated to the ESM2 breakage report." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "sync_gpus kwargs update bug and chunked_prefill cache-missing bug are distinct generation paths." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "processor.save_pretrained missing files is unrelated to deepspeed causing optimizer.step to run twice." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Data filtering slowdown and double optimizer step are different failures." + }, + { + "left": "issue:43673", + "right": "issue:43827", + "accept": false, + "reason": "Generation cache regression is unrelated to docs still referencing removed pipeline APIs." + }, + { + "left": "issue:43701", + "right": "issue:44749", + "accept": false, + "reason": "resume_from_checkpoint key mismatch is unrelated to the filtering slowdown." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Gemma eos_token_id mismatch is a config bug, not an AttentionInterface.register behavior change." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "A generic bug report is not the same concrete issue as gemma-4 zero3 from_pretrained." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "SAM3 video encoder missing fpn_position_embeddings is unrelated to qwen2.5-vl video position_id logic." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure and Qwen2.5-GGUF loading failure are different operations and failure modes." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "SAM3 VisionEncoderOutput field omission is unrelated to AutoConfig trust_remote_code missing fields." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "PyTorch missing ImportError and Sentence Transformers hanging on load are not the same bug." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe breakage in transformers 5.0 is unrelated to processor.save_pretrained missing files." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Gemma4 attention-mask bug and MaskGenerationPipeline partial-batch bug are different code paths." + }, + { + "left": "issue:45406", + "right": "issue:46202", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer in serve is unrelated to Gemma pipeline roundtrip failure on \u2581." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 API guidance is unrelated to InternVL2 meta-initialization incompatibility." + }, + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Cannot import AutoImageProcessor from transformers is a different import issue than a nonexistent image_utils import." + }, + { + "left": "issue:43475", + "right": "issue:44661", + "accept": false, + "reason": "SAM3 output attribute bug is unrelated to add-new-model-like failing inside TOKENIZER_MAPPING_NAMES." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized saving failure is unrelated to Llama4Config missing pad_token_id." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but config-field dropping and from_pretrained error are not the same concrete bug." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Mistral regex patch crash and auto_docstring annotation crash are different AttributeErrors in different subsystems." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "Both mention Qwen3.5, but one is a from_pretrained error and the other is a missing config field being dropped." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "Quantization offloading issues are unrelated to incorrect output under pipeline parallelism." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs sync bug is unrelated to attention-kernel divergence 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": 38, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "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", + "issue:43232", + "issue:43278", + "issue:43377", + "issue:43479", + "issue:43504", + "issue:43526", + "issue:43673", + "issue:43704", + "issue:43723", + "issue:43784", + "issue:43824", + "issue:43976", + "issue:44038", + "issue:44079", + "issue:44188", + "issue:44222", + "issue:44242", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44792", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:44964", + "issue:45103", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45561", + "issue:45712", + "issue:45753", + "issue:45812" + ], + "soft_pairs": [ + "issue:41628|issue:44351", + "issue:42673|issue:43704", + "issue:44945|issue:45753", + "issue:44964|issue:45137", + "issue:36010|issue:41628", + "issue:42831|issue:44222", + "issue:42994|issue:43824", + "issue:44222|issue:45310", + "issue:43673|issue:44188", + "issue:43232|issue:44792", + "issue:44038|issue:45310", + "issue:42175|issue:45712", + "issue:43526|issue:45561", + "issue:44496|issue:45561", + "issue:43504|issue:45561", + "issue:42175|issue:44933", + "issue:43377|issue:44242", + "issue:42898|issue:44568", + "issue:43976|issue:44933", + "issue:42222|issue:45200", + "issue:30990|issue:35141", + "issue:43278|issue:45137", + "issue:43479|issue:44869", + "issue:43479|issue:44079", + "issue:44623|issue:45310", + "issue:30333|issue:33290", + "issue:42898|issue:45812", + "issue:43784|issue:43824", + "issue:44734|issue:45103", + "issue:43723|issue:44568", + "issue:42898|issue:44779" + ], + "prepared_review_unit_hash": "ebb691b9a5cba425c7a93aa3884c8a71b4295484a152a3322428e0a653fa42de", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36849, + "estimated_input_tokens": 9213, + "estimated_eval_tokens": 18682 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of distinct Transformers regressions (imports, tokenizer behavior, model loading/saving, multimodal bugs, training/runtime issues). Only one soft edge looks like a true duplicate: the two Qwen3ForCausalLM VRAM-leak reports.", + "confidence": 0.87, + "canonical_issue_reason": "issue:42673 is the clearest canonical issue because it is the fuller, earlier report of the same Qwen3ForCausalLM multi-dataloader-thread VRAM leak that 43704 appears to duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42673 is the best issue anchor in the cluster: it has the strongest duplicate evidence, a detailed repro, and the paired report 43704 looks like the same bug.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols (`AutoImageProcessor` vs `HybridCache`) and different code paths." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same title, same model, same symptom; this looks like the same Qwen3ForCausalLM VRAM-leak bug reported twice." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Different failures: incorrect LLM output under pipeline parallelism vs an infinite loop on a specific image for Qwen3_5." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Unrelated issues: Phi-4 multimodal model loading vs DeepSpeed ZeRO3 deque underflow." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "Both are import regressions, but they are different missing names and likely separate fixes." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "Different bugs: FP8 accuracy drift vs FP8 `save_pretrained` failure in MoE." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized model saving and missing `Qwen2_5_VLForConditionalGeneration` import are separate problems." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "Both mention FP8/MoE, but one is save-time behavior and the other is a `from_pretrained` load error for a different Qwen model variant." + }, + { + "left": "issue:43673", + "right": "issue:44188", + "accept": false, + "reason": "Different subsystems and symptoms: generation cache handling vs attention-kernel divergence under `torch.compile`." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Different concrete bugs: generation kwargs after `sync_gpus` vs a janus image-generation test failure." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both involve Qwen MoE models, but the model variants and failure modes differ; too broad to treat as the same bug." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Packaging/backend-dependency issue vs leftover dummy classes leaking into the module namespace." + }, + { + "left": "issue:43526", + "right": "issue:45561", + "accept": false, + "reason": "Image-processor label reduction bug vs xdist race on a captured-info file are unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Model config loading failure vs test-worker file race; no shared code path." + }, + { + "left": "issue:43504", + "right": "issue:45561", + "accept": false, + "reason": "Beit legacy-field load bug vs xdist test race are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "TensorFlow extra/backend packaging issue vs a missing `image_utils` import; different failure classes." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "MIMI padding-mask mismatch vs MoE load-balancing loss omission are different model bugs." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but one concerns cleanup spaces and the other special-token insertion for a specific tokenizer." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version compatibility problem vs missing import from `image_utils`; not the same bug." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Different models and symptoms: broken ViTPose models vs Gemma 4 missing default `mm_token_type_ids`." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "Sentence-Transformers loading hang and embedding reinitialization on resize are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "Embedding dtype drift during train/eval vs DeepSpeed ZeRO3 deque underflow are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Phi-4 multimodal config defaults vs Whisper timestamp decoding crash are different code paths." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Config initialization with `None` and ModelOutput key assignment are separate abstractions and bugs." + }, + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Processor save-pretrained file emission and Qwen3.5 Moe load errors are different operations and likely different fixes." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting on failed training jobs vs Adafactor OOM in DeepSpeed are unrelated runtime issues." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions, but the affected models and concrete breakages differ." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Sentence-Transformers import failure due to `nn` vs missing export of a Qwen VL class are unrelated import problems." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "Serving KV-cache indexing crash and auto-docstring annotation handling are unrelated." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Tokenizer loading failure in v5 vs special-token insertion regression for mdeberta-v3-base are different bugs." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but one is a global v5 cleanup-space change and the other is a DeepSeek tokenizer correctness bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays mostly conservative. The one accepted soft edge (issue:42673 \u2194 issue:43704) is supported by identical titles and matching symptoms. The rejected soft edges are framed cautiously and do not overclaim duplicates. One minor wording point: the packet does not explicitly prove a \u2018detailed repro,\u2019 but the fuller body length and discussion activity make the anchor choice reasonable enough 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": 39, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:22355", + "issue:28282", + "issue:30333", + "issue:34689", + "issue:35141", + "issue:36331", + "issue:41628", + "issue:42757", + "issue:42831", + "issue:42898", + "issue:43116", + "issue:43122", + "issue:43232", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43761", + "issue:43874", + "issue:44060", + "issue:44190", + "issue:44206", + "issue:44222", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44448", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44623", + "issue:44933", + "issue:44977", + "issue:45071", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45245", + "issue:45310", + "issue:45356", + "issue:45405", + "issue:45406", + "issue:45656", + "issue:45712", + "issue:45834", + "issue:46082", + "issue:46132", + "issue:46209" + ], + "soft_pairs": [ + "issue:42898|issue:45356", + "issue:43874|issue:45092", + "issue:45405|issue:45712", + "issue:44060|issue:45125", + "issue:42898|issue:44448", + "issue:44295|issue:45656", + "issue:44448|issue:46132", + "issue:44933|issue:45712", + "issue:44222|issue:44297", + "issue:44297|issue:45656", + "issue:43116|issue:44190", + "issue:43479|issue:43644", + "issue:44190|issue:46082", + "issue:44222|issue:44623", + "issue:30333|issue:34689", + "issue:42831|issue:44623", + "issue:43122|issue:46209", + "issue:30333|issue:36331", + "issue:22355|issue:28282", + "issue:45245|issue:46132", + "issue:28282|issue:30333", + "issue:30333|issue:35141", + "issue:45071|issue:46082", + "issue:41628|issue:42757", + "issue:44206|issue:45103", + "issue:44297|issue:45310", + "issue:43723|issue:44977", + "issue:43232|issue:45834", + "issue:44263|issue:45092", + "issue:43761|issue:46082", + "issue:44222|issue:44484", + "issue:44373|issue:44485", + "issue:44295|issue:44623", + "issue:44483|issue:45406" + ], + "prepared_review_unit_hash": "54dc6f678cd45e18622c9327e613bfe8eace4f9bad4ac5d8fdc95cb649560203", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37700, + "estimated_input_tokens": 9425, + "estimated_eval_tokens": 19106 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These are mostly keyword-similar but substantively different issues; none of the soft pairs look like true duplicates.", + "confidence": 0.95, + "canonical_issue_reason": "No single canonical issue stands out: the set mixes unrelated import errors, model regressions, config validation bugs, and save/load edge cases.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45356 is the strongest standalone report here because it has a clear regression boundary, affected model, and concrete symptom, but it still should not absorb the rest of the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer regressions, but they describe different failures: v5 clean_up_tokenization_spaces behavior vs Kimi-K2.5 codec handling and warning text." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "Different multimodal bugs: one is a missing image-patch API, the other is a remote-code checkpoint/meta-init incompatibility." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "Unrelated: one is a dependency version bump problem, the other is cleanup of leftover dummy classes." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "Both are Qwen-related, but one is a tied-weights warning and the other is missing tensor-parallel metadata; different code paths." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Both mention v5 regressions, but they affect different models and different behavior paths." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "A buffer read error and a double optimizer step under DeepSpeed are unrelated bugs." + }, + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "Different problems: Pegasus output regression versus changed behavior in AttentionInterface.register." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "Import/export failure in image_utils is unrelated to leftover dummy class cleanup." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "Both involve save_pretrained, but one is FP8 MoE serialization and the other is tokenizer config metadata mismatch." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save metadata and optimizer stepping are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Both are example-script reports, but one is multi-label classification output and the other is local dataset loading." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Different subsystems: multimodal config defaults versus non-persistent buffer initialization in v5." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Example dataset loading failure is unrelated to LlamaConfig head_dim validation." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Both touch save_pretrained, but one is model FP8/MoE behavior and the other is processor file emission." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow job reporting on failure is unrelated to a model-loading regression." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FP8 accuracy issues and missing processor save files are different bugs." + }, + { + "left": "issue:43122", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer differences across versions are unrelated to FP8 inference failures on B200." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "A failed-job reporting bug and a Trainer compute_loss signature change are not the same issue." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Both are import errors, but one is a missing transformers.onnx module and the other is a missing PyTorch dependency." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category-count runtime error and AttentionInterface.register behavior change are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "Dependency import failure versus MLflow reporting bug; no shared underlying cause." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "Job-status reporting and embedding reinitialization are separate problems." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "Both are config-related, but one is type checking regression and the other is explicit head_dim validation." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Different import/export issues in different packages, with different fixes." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "Unsupported processor argument and auto_docstring annotation crash are unrelated." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer save metadata mismatch is unrelated to Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "Tokenizer loading and flash-attention generation issues affect different code paths." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs synchronization and OCR index overflow are unrelated." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split return handling in GlmMoeDsaIndexer is unrelated to meta-initialization incompatibility." + }, + { + "left": "issue:43761", + "right": "issue:46082", + "accept": false, + "reason": "CLIP hidden_states regression and LlamaConfig validation are different bugs." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 save_pretrained bug and a question about max_shard_size are not duplicates." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "Wrong docstring and RoPE implementation discussion do not describe the same issue." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading a registered buffer and missing processor save files are unrelated." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat-completions request handling and Gemma4Processor missing _tokenizer are different failures." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 40, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:30990", + "issue:38175", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42757", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43232", + "issue:43479", + "issue:43576", + "issue:43646", + "issue:43723", + "issue:43756", + "issue:43784", + "issue:43824", + "issue:44038", + "issue:44060", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44360", + "issue:44373", + "issue:44442", + "issue:44514", + "issue:44568", + "issue:44671", + "issue:44704", + "issue:44829", + "issue:44871", + "issue:44913", + "issue:44991", + "issue:45092", + "issue:45103", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45593", + "issue:45753", + "issue:46002", + "issue:46050", + "issue:46132" + ], + "soft_pairs": [ + "issue:44263|issue:44360", + "issue:44514|issue:45103", + "issue:45092|issue:45357", + "issue:43723|issue:44871", + "issue:30990|issue:38175", + "issue:42907|issue:46050", + "issue:44261|issue:45092", + "issue:43723|issue:44991", + "issue:44829|issue:46002", + "issue:43723|issue:44442", + "issue:41628|issue:43824", + "issue:42831|issue:45310", + "issue:43479|issue:44913", + "issue:44263|issue:44373", + "issue:42831|issue:44038", + "issue:39401|issue:44568", + "issue:42757|issue:45278", + "issue:43232|issue:45593", + "issue:42886|issue:42994", + "issue:42886|issue:43066", + "issue:22355|issue:28218", + "issue:42831|issue:45753", + "issue:43756|issue:44060", + "issue:44117|issue:44704", + "issue:44671|issue:46132", + "issue:43576|issue:43784", + "issue:43646|issue:46002", + "issue:42371|issue:44263" + ], + "prepared_review_unit_hash": "501ad87e7f6f3f9299b96610247ce7c58513db3d00f5bfe33bb485cacbc2eb41", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35135, + "estimated_input_tokens": 8784, + "estimated_eval_tokens": 17824 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a grab-bag of unrelated Transformers regressions; the soft-similarity pairs share broad themes (tokenizers, import errors, save/load, multimodal configs) but not the same concrete bug. No soft edge is safe to merge.", + "confidence": 0.87, + "canonical_issue_reason": "issue:43723 is the best umbrella representative because it is a broad, user-facing tokenizer-loading regression in v5 and is closest to several other tokenizer/import complaints without being overly narrow.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43723 is the strongest global representative for the cluster: it captures the recurring tokenizer/from_pretrained failure mode that appears most central here, while the other issues are mostly narrower or unrelated variants.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Different DSA-related bugs: one is about torch.split return values, the other about a missing ReLU. Same subsystem, not the same defect." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "Completely different code paths: Qwen2.5-VL chat template batching versus auto_docstring annotation handling." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "Both are multimodal model issues, but one is meta-init incompatibility for old remote-code checkpoints and the other is incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Tokenizer-loading regression versus an eos_token_id configuration mismatch; related area, but different bug." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Different models and symptoms: loading gets stuck versus zero probabilities in SigLIP2." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Both involve quantized model persistence, but they target different quantization flows and failure modes." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "Layernorm precision/config issue is unrelated to meta-initialization compatibility for remote-code checkpoints." + }, + { + "left": "issue:43723", + "right": "issue:44991", + "accept": false, + "reason": "Both mention tokenizer loading, but the titles point to different models and likely different root causes." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Flash-attention training degradation and NaNs in classification heads are different failures, not one bug." + }, + { + "left": "issue:43723", + "right": "issue:44442", + "accept": false, + "reason": "AutoTokenizer failing on one tokenizer class is not the same bug as the broader v5 tokenizer-loading regression." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different export/mapping causes." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue is unrelated to Qwen3.5 MoE from_pretrained failure." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Different configuration bugs in different model families: default multimodal configs versus rotary_pct reload behavior." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "One concerns torch.split return values, the other a docstring error; not the same bug." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "Different Qwen/FP8 versus Qwen3-VL-Moe issues with no concrete shared code-path." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Offset_mapping behavior and BOS/EOS special-token insertion are distinct tokenizer bugs." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "Missing is_offline_mode import is not the same as a broad set of import errors after upgrading to v5.5.0." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation kwargs sync issue and D-FINE auxiliary-loss behavior are unrelated." + }, + { + "left": "issue:42886", + "right": "issue:42994", + "accept": false, + "reason": "Offline tokenizer cache loading and quantized model saving are different persistence problems." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Both concern tokenizers, but one is offline cache loading and the other is decoder-type configuration in v5." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "No module import error for transformers.onnx is unrelated to added-token spacing behavior." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FP8 accuracy regression and an infinite-generation loop on a specific image are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "RoPE-layer omission and tied-weights warning are different model-architecture bugs." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES None handling and AutoProcessor kwargs propagation are separate from_pretrained issues." + }, + { + "left": "issue:44671", + "right": "issue:46132", + "accept": false, + "reason": "CamemBERT prediction regression is unrelated to AttentionInterface.register behavior changes." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "Broken env command and sentence-transformers import NameError are different regressions." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage is not the same as NaNs in classification heads." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 settings misuse is unrelated to the DSA indexer split/ReLU 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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:29159", + "issue:30990", + "issue:33357", + "issue:36296", + "issue:39401", + "issue:41628", + "issue:42757", + "issue:42907", + "issue:43116", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43784", + "issue:43824", + "issue:43906", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44188", + "issue:44222", + "issue:44261", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44373", + "issue:44485", + "issue:44488", + "issue:44514", + "issue:44521", + "issue:44625", + "issue:44704", + "issue:44743", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45200", + "issue:45216", + "issue:45278", + "issue:45290", + "issue:45446", + "issue:45478", + "issue:45600" + ], + "soft_pairs": [ + "issue:43122|issue:44521", + "issue:30990|issue:33357", + "issue:44295|issue:44297", + "issue:44188|issue:44373", + "issue:44704|issue:45290", + "issue:41628|issue:45278", + "issue:42907|issue:44945", + "issue:44261|issue:44373", + "issue:43232|issue:44743", + "issue:43784|issue:45600", + "issue:39401|issue:44488", + "issue:43976|issue:44945", + "issue:43408|issue:43475", + "issue:42907|issue:43937", + "issue:43122|issue:44514", + "issue:39401|issue:44521", + "issue:44360|issue:44485", + "issue:44625|issue:44829", + "issue:42907|issue:44964", + "issue:43116|issue:45200", + "issue:44038|issue:44297", + "issue:44038|issue:44295", + "issue:28218|issue:30990", + "issue:43122|issue:44779", + "issue:42907|issue:43452", + "issue:36296|issue:43116", + "issue:43824|issue:44351", + "issue:44038|issue:44222", + "issue:45216|issue:45478", + "issue:43122|issue:43906", + "issue:43784|issue:45446", + "issue:42757|issue:44351", + "issue:22355|issue:29159" + ], + "prepared_review_unit_hash": "6049652fbcc15fc4679699e7d40e2dbcd64c14d11ff0d5fad82717484e954248", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37204, + "estimated_input_tokens": 9301, + "estimated_eval_tokens": 18858 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: most items are unrelated regressions or import/model issues that only share a broad subsystem. I would anchor on issue:43122 as the most representative tokenizer-regression report, but I do not see any soft pair that is clearly the same concrete bug/change.", + "confidence": 0.68, + "canonical_issue_reason": "issue:43122 is the broadest and most central tokenizer-regression report here, and it overlaps most with the other tokenizer-related items in the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "It is the best representative issue for the tokenizer-regression slice of this mixed cluster: specific enough to be actionable, but broad enough to cover the nearby tokenization complaints.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization/chat-template behavior, but one is a generic tokenizer regression and the other is multimodal assistant_mask generation; different defect." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Different failures: Sentence-Transformers loading hang vs a Mac bus error in CLIP model loading." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Unrelated symptoms: position_ids buffer access error vs tokenizer_class mismatch in tokenizer_config.json." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence under torch.compile is unrelated to a wrong docstring for position_ids." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "Different code paths: AutoProcessor kwargs/cached_file plumbing vs apply_chat_template crashing on tool-call messages." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-related, but one is a missing AutoImageProcessor export and the other is a broader upgrade breakage; not the same bug." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Saving dequantized models and incorrect pipeline-parallel outputs are different problems." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "Precision/config mismatch in MLA layernorm vs a docstring issue; no duplicate signal." + }, + { + "left": "issue:43232", + "right": "issue:44743", + "accept": false, + "reason": "Both touch generation internals, but sync_gpus kwarg handling and Qwen3.5 recurrent-state resets are distinct defects." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import NameError in sentence-transformers integration vs stale auto_mappings references; different breakages." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Qwen3 offset_mapping is unrelated to a model-loading failure for cjvt/sleng-bert." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version support regression is unrelated to incorrect outputs under pipeline parallelism." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "Both are SAM3-related, but a model-type warning and a missing attribute on vision encoder output are not the same bug." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Dequantized save failure vs invalid GenerationConfig for GLM-5 are different issues." + }, + { + "left": "issue:43122", + "right": "issue:44514", + "accept": false, + "reason": "Tokenizer version-drift regression vs batched apply_chat_template crash with padding=False; different concrete symptoms." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Different multimodal/tokenizer bugs: wrong offset_mapping vs assistant_masks all zero." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "DSA indexer/ReLU discussion is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and flash_attention_3 training degeneration are different code paths." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Saving dequantized checkpoints is unrelated to a model-load failure for Phi-4 multimodal instruct." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Example-script multi-label empties and Gemma 4 mm_token_type_ids defaulting are different bugs." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe issue vs tokenizer_class mismatch; no shared concrete defect." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe bug is unrelated to the position_ids buffer access problem." + }, + { + "left": "issue:28218", + "right": "issue:30990", + "accept": false, + "reason": "Tokenizer spacing bug vs Sentence-Transformers loading hang; unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but one is broad version drift and the other is Deepseek-specific; not enough evidence of the same root cause." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Dequantized save_pretrained failure vs gguf_file loading breakage are different artifacts." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training bug and example-script classification output bug are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Different missing symbols: Qwen2_5_VLForConditionalGeneration vs HybridCache." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "Both mention MoE, but one is a model-specific loading/behavior issue and the other is FP8 save_pretrained; different fixes." + }, + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Related Qwen3.5 family, but save_pretrained producing a bad checkpoint and from_pretrained throwing an error are not clearly the same concrete bug." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "Importing sentence-transformers NameError and a PyTorch version check in flex_attention are unrelated." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Missing is_offline_mode from huggingface_hub and missing HybridCache from transformers are different import failures." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "No module named transformers.onnx is unrelated to slow-vs-fast tokenizer inconsistency." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:33453", + "issue:38617", + "issue:39401", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:43064", + "issue:43066", + "issue:43116", + "issue:43278", + "issue:43441", + "issue:43493", + "issue:43611", + "issue:43673", + "issue:43825", + "issue:43827", + "issue:43874", + "issue:43906", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44117", + "issue:44291", + "issue:44380", + "issue:44485", + "issue:44704", + "issue:44779", + "issue:44936", + "issue:44938", + "issue:44945", + "issue:45071", + "issue:45072", + "issue:45083", + "issue:45278", + "issue:45357", + "issue:45446", + "issue:45478", + "issue:45563", + "issue:45588", + "issue:45753", + "issue:46002", + "issue:46077" + ], + "soft_pairs": [ + "issue:43278|issue:44945", + "issue:42175|issue:44938", + "issue:42673|issue:45072", + "issue:43493|issue:45753", + "issue:38617|issue:45278", + "issue:43066|issue:44779", + "issue:43992|issue:44704", + "issue:22355|issue:29127", + "issue:43611|issue:46002", + "issue:43937|issue:44485", + "issue:43976|issue:45446", + "issue:42222|issue:43441", + "issue:42371|issue:43874", + "issue:43116|issue:43493", + "issue:44075|issue:44936", + "issue:45357|issue:45478", + "issue:44038|issue:45478", + "issue:43064|issue:43673", + "issue:43116|issue:43825", + "issue:45083|issue:45446", + "issue:43937|issue:45563", + "issue:44380|issue:46077", + "issue:44117|issue:45588", + "issue:33453|issue:39401", + "issue:43827|issue:45446", + "issue:39401|issue:44779", + "issue:42673|issue:43906", + "issue:44291|issue:45071" + ], + "prepared_review_unit_hash": "f43a8972b49992175556c075559720b0bac1bbe746631f45409a3bb8848a309a", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35172, + "estimated_input_tokens": 8793, + "estimated_eval_tokens": 17842 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is thematically noisy but mostly centers on Transformers v5 regressions, tokenizer/loading issues, and model-specific bugs. The soft pairs do not look like true duplicates under a strict code-path standard, so all candidate edges are rejected.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45278 is the broadest umbrella report here: it captures widespread v5 import breakage after upgrade, making it the closest cluster anchor.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best overall representative because it is the most general and high-impact regression report in the set, though it is not a safe duplicate sink for the more specific issues.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Different bugs: dtype drift between train/eval vs incorrect output under pipeline parallelism." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Packaging/backend dependency issue vs Python 3.14 import/load failure." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in threaded dataloader use vs bfloat16 dtype mismatch in inference." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "SigLIP2 implementation discrepancy vs Qwen3_5 infinite loop on a specific image." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "Both are import-related, but one is a specific missing symbol and the other is a broad upgrade report; not the same concrete failure." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Tokenizer decoder type regression vs Deepseek tokenizer output corruption; related area, distinct bug." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Missing model weight on load vs AutoProcessor not forwarding kwargs to cached_file." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Missing module import vs LayoutLMv3 error-message clarity; unrelated." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Model loading breakage from base_model_prefix vs NaNs in classification heads after init/checkpoint." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "GenerationConfig validation failure vs GLM-5 RoPE implementation discussion; not the same code path." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python support/version compatibility issue vs PyTorch version guard for AuxRequest import." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "VitPose model breakage vs Ministral-3 FlashAttention failure; different models and failures." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 behavior settings vs missing image-patch helper attribute; unrelated." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Example script returns empty multi-label results vs SigLIP2 model discrepancy." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer argument handling bug vs trainer.evaluate failure after training; different subsystems." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both concern Qwen3.5, but save_pretrained key layout and from_pretrained loading are different concrete bugs." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Related model family, but Qwen3-VL-Moe and Qwen3.5-Moe load failures are not the same underlying bug." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Distributed optimizer-state corruption vs missing GenerationMixin cache in chunked prefill." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "Classification example misbehavior vs pipeline translation support/error-message issue." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Qwen3_omni_moe helper behavior vs PyTorch version check; unrelated." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "Invalid generation config vs stale num_return_sequences warning; different generation-path defects." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "GPT2 attention scaling ignored vs Gemma4 bidirectional-attention mask construction; different model-specific behaviors." + }, + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "Tokenizer mapping None handling vs flash_attention crash on s_aux=None; unrelated code paths." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Tokenizer loading regression vs Qwen3 offset_mapping bug; both tokenizer-related but not the same failure." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() vs a runtime version-check bug; unrelated." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Different tokenizer bugs for different model families and symptoms; not one duplicate." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "VRAM leak in dataloader threading vs isolated reproduction of a different issue; no concrete match." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Unexpected init_empty_weights argument vs PretrainedConfig type-checking regression; distinct loading/type 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": 43, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42490", + "issue:42617", + "issue:42673", + "issue:42831", + "issue:43012", + "issue:43066", + "issue:43122", + "issue:43408", + "issue:43421", + "issue:43450", + "issue:43577", + "issue:43582", + "issue:43673", + "issue:43742", + "issue:43937", + "issue:43986", + "issue:44075", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44410", + "issue:44466", + "issue:44483", + "issue:44484", + "issue:44496", + "issue:44829", + "issue:44933", + "issue:44977", + "issue:45081", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45405", + "issue:45446", + "issue:45464", + "issue:45478", + "issue:45542", + "issue:45561", + "issue:45656", + "issue:46002", + "issue:46178" + ], + "soft_pairs": [ + "issue:43450|issue:43986", + "issue:43937|issue:44977", + "issue:44466|issue:46002", + "issue:45405|issue:45446", + "issue:44117|issue:45081", + "issue:45341|issue:45561", + "issue:38617|issue:41628", + "issue:42222|issue:46178", + "issue:44410|issue:45310", + "issue:42617|issue:45341", + "issue:44829|issue:45216", + "issue:39401|issue:43421", + "issue:42673|issue:43582", + "issue:42490|issue:42673", + "issue:43408|issue:43742", + "issue:42371|issue:44261", + "issue:44933|issue:45278", + "issue:38617|issue:44220", + "issue:44484|issue:45310", + "issue:43937|issue:44496", + "issue:43066|issue:43122", + "issue:42222|issue:45542", + "issue:44483|issue:45464", + "issue:42371|issue:42831", + "issue:42175|issue:42222", + "issue:44075|issue:45656", + "issue:43673|issue:45341", + "issue:44410|issue:45478", + "issue:43012|issue:43577" + ], + "prepared_review_unit_hash": "e6d2d338ad7c6589274def7a8ad2d0bb476f8addc246a7ca4c57ac5b06e1ce31", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36161, + "estimated_input_tokens": 9041, + "estimated_eval_tokens": 18338 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: most items are unrelated single-bug reports, and none of the soft pairs look like true duplicates or the same concrete change. I would not collapse any of the candidates together.", + "confidence": 0.94, + "canonical_issue_reason": "No single canonical issue: the cluster spans separate import/export regressions, tokenizer behavior changes, model-loading errors, backend/config problems, and optimizer/runtime bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue is a strong global representative because the soft pairs do not share one underlying code-path problem.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Different bugs: batched video processor shape handling vs missing torchvision causing AutoProcessor load failure." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "GenerationConfig validation error and Qwen3.5 flash-attention generation failure are separate problems." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "Serialization/tied-weight inconsistency is unrelated to NaNs in classification heads after checkout/init." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A release/version pin issue for PEFT is not the same as a PyTorch version check bug in flex_attention." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Generic tokenizer mapping lookup failure and Mistral regex patch crash affect different code paths." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "A minor testing_utils bug is not the same as an xdist race on captured_info.txt." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "These are different missing-import issues for different symbols and modules." + }, + { + "left": "issue:42222", + "right": "issue:46178", + "accept": false, + "reason": "Broken VitPose models and tensorflow_text backend mapping errors are unrelated." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing Qwen3Next projections and Qwen3.5 MoE from_pretrained errors are different model-specific failures." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "A script execution problem is unrelated to a testing_utils bug." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "flash_attention_3 training degeneration and Qwen3.5 checkpoint serialization are different issues." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Wrong offset_mapping in Qwen3 tokenizer is not the same as post-processor not updating after runtime special-token changes." + }, + { + "left": "issue:42673", + "right": "issue:43582", + "accept": false, + "reason": "VRAM leak in multi-threaded dataloaders is unrelated to an Apple Silicon TypeError in caching_allocator_warmup." + }, + { + "left": "issue:42490", + "right": "issue:42673", + "accept": false, + "reason": "Local-path import warning and VRAM leak are different symptoms and code paths." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "sam3_video/sam3_tracker config mismatch is unrelated to the MobileLLM key error." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API guidance/config warning and missing rms_norm_eps precision error are distinct problems." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "A nonexistent image_utils import and broad import errors after upgrade are related in theme but not the same concrete bug." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "ImportError for layer_type_validation is unrelated to the _torch_extract_fbank_features() issue." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "save_pretrained shard-size behavior and Qwen3.5 MoE loading error are different code paths." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "GenerationConfig invalidity is not the same as an unrecognized model missing model_type." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Wrong decoder type in v5 and changed tokenization across versions are related to tokenizers but not the same underlying bug." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "VitPose model breakage is unrelated to a backend error from installing tensorboard without TensorFlow." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "chat/completions request rejection and Qwen3.5 streaming failure are different failures in the API stack." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 behavior warning is unrelated to the FineGrainedFP8 accuracy issue." + }, + { + "left": "issue:42175", + "right": "issue:42222", + "accept": false, + "reason": "TensorFlow backend missing from a torch install and broken VitPose models are unrelated." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "SGD args being ignored and optimizer.step being called twice are different optimizer bugs." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Generation cache missing in chunked_prefill is unrelated to a small testing_utils bug." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Both mention Qwen3.5/Qwen3Next, but the missing projections bug and the from_pretrained error are not the same concrete failure." + }, + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "A PyTorch compile warning on bfloat16 models is unrelated to Blip2 dtype fields remaining float32 on load." + } + ] + }, + "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": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:38617", + "issue:41628", + "issue:42757", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43335", + "issue:43408", + "issue:43441", + "issue:43452", + "issue:43502", + "issue:43531", + "issue:43673", + "issue:43824", + "issue:43873", + "issue:43874", + "issue:43931", + "issue:43986", + "issue:43992", + "issue:44038", + "issue:44062", + "issue:44117", + "issue:44261", + "issue:44263", + "issue:44315", + "issue:44351", + "issue:44360", + "issue:44368", + "issue:44410", + "issue:44484", + "issue:44485", + "issue:44493", + "issue:44704", + "issue:44829", + "issue:44987", + "issue:45003", + "issue:45071", + "issue:45103", + "issue:45341", + "issue:45372", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45542", + "issue:45588", + "issue:46050", + "issue:46123", + "issue:46172" + ], + "soft_pairs": [ + "issue:43097|issue:44493", + "issue:41628|issue:45071", + "issue:43874|issue:44261", + "issue:44263|issue:44485", + "issue:45071|issue:45542", + "issue:44263|issue:46172", + "issue:44360|issue:46172", + "issue:43012|issue:43408", + "issue:43441|issue:44829", + "issue:44484|issue:45478", + "issue:43931|issue:44410", + "issue:43335|issue:45071", + "issue:43931|issue:44038", + "issue:43673|issue:46123", + "issue:43097|issue:43452", + "issue:45003|issue:45341", + "issue:43986|issue:45341", + "issue:44117|issue:44987", + "issue:43408|issue:44368", + "issue:43874|issue:44263", + "issue:43986|issue:44484", + "issue:45440|issue:46172", + "issue:43531|issue:45341", + "issue:43986|issue:45372", + "issue:44261|issue:44485", + "issue:38617|issue:42757", + "issue:43097|issue:46050", + "issue:43097|issue:45588", + "issue:43064|issue:44315", + "issue:44062|issue:44117", + "issue:44410|issue:45375", + "issue:38617|issue:44351", + "issue:44038|issue:44410", + "issue:43502|issue:43992", + "issue:43097|issue:43873", + "issue:44704|issue:45103", + "issue:43986|issue:44117", + "issue:43824|issue:43986" + ], + "prepared_review_unit_hash": "f824affe9cf75f0f1992ce55c6cc1a61142bf34f0bddbb5cd72cb7283bb99527", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 38, + "serialized_chars": 39074, + "estimated_input_tokens": 9769, + "estimated_eval_tokens": 19794 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No safe duplicates stand out here: the soft pairs mostly share broad subsystem keywords (imports, Qwen, flash attention, warnings) but describe different concrete failures. I would keep issue 43873 as the best representative standalone report in the set.", + "confidence": 0.79, + "canonical_issue_reason": "issue:43873 is the strongest standalone bug report in the batch: open, detailed, and the most actively discussed/inbound-referenced item here.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 has the clearest ongoing impact and enough activity to serve as the most useful representative issue for this collection.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Both mention config/model warnings, but one is about a removed config field and the other about unexpected position-id keys; different bug reports." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "Different failures: missing top-level import vs. PretrainedConfig type-checking regression." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different model families and code paths: image-patch helper missing method vs. MLA layernorm epsilon precision issue." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "Both are GLM-related, but one is a DSA indexer split bug and the other is a RoPE implementation discussion; not the same defect." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "PretrainedConfig type checking and TensorBoard/TF backend detection are unrelated bugs." + }, + { + "left": "issue:44263", + "right": "issue:46172", + "accept": false, + "reason": "Completely different model/code paths: GlmMoeDsaIndexer split behavior vs. DeepSeek-V4 tensor transpose logic." + }, + { + "left": "issue:44360", + "right": "issue:46172", + "accept": false, + "reason": "A missing ReLU in one GLM indexer path is unrelated to the DeepSeek-V4 comb transpose bug." + }, + { + "left": "issue:43012", + "right": "issue:43408", + "accept": false, + "reason": "Both are warnings, but they concern different models and different warning conditions." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both involve flash attention, but one is a Ministral-3 load failure and the other is degenerate training in sequence classification; not mergeable as one fix." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "Save shard-size behavior is unrelated to the Qwen3.5-MoE from_pretrained loading error." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Both are Qwen model issues, but one is a Qwen3-VL weight-shape mismatch and the other is missing projections in qwen3next; different bugs." + }, + { + "left": "issue:43335", + "right": "issue:45071", + "accept": false, + "reason": "A SwitchTransformers config construction bug is unrelated to a PretrainedConfig typing regression." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both mention Qwen3-VL variants, but one is a specific shape mismatch and the other is a broader v5/Qwen3-VL-Moe compatibility report." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Generation cache missing in chunked prefill and mask pipeline final-batch dropping are different subsystems and symptoms." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Removed config field warnings and GGUF tokenizer/model loading failures are unrelated." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "modeling_utils sys.modules access and a testing_utils bug do not describe the same failure." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "AutoProcessor/torchvision crash and a testing_utils issue are unrelated." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "A tokenizer mapping None-handling bug is not the same as a model-specific load failure for physical-intelligence/fast." + }, + { + "left": "issue:43408", + "right": "issue:44368", + "accept": false, + "reason": "Different warnings in different contexts: SAM3 model-type mismatch vs. tie_word_embeddings guidance for Qwen fine-tuning." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Different GLM subcomponents and different failures; no shared underlying code-path bug is evident." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "Missing torchvision error in AutoProcessor is unrelated to shard-size behavior in save_pretrained." + }, + { + "left": "issue:45440", + "right": "issue:46172", + "accept": false, + "reason": "Both are DeepSeek-related, but one is a model implementation divergence and the other is a specific tensor operation bug." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and a testing_utils bug are unrelated." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Video-model loading without torchvision and Gemma 4 processor loading without ReasoningEffort are different import/runtime failures." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "Layernorm epsilon precision issue and GLM-5 RoPE implementation are not the same bug." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they are for different missing symbols in different packages and do not share a single fix." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "A removed config field warning and a quantization save/load breakage are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Tie-embeddings warning/reporting and flash_attention s_aux=None crash are different issues." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Optimizer state corruption under FSDP2/PEFT is unrelated to Liger Kernel application with model_init." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "AddedToken constructor argument collision and TOKENIZER_MAPPING_NAMES None-handling are different tokenizer-loading bugs." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Different Qwen configuration problems: missing attention projections vs. a silently dropped vision config field." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different APIs/symbols; no evidence they stem from the same regression." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Qwen3-VL-Moe loading trouble and qwen3next missing projections are distinct model-specific defects." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "Local-files-only network leakage and missing embed_tokens weight on UMT5 loading are unrelated." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "A removed config field warning is unrelated to quantization/offloading behavior." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "AutoProcessor kwargs forwarding and _process_kwargs_parameters crashing under future annotations are different issues." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Video processor loading without torchvision and tokenizer mapping None-handling are unrelated." + }, + { + "left": "issue:43824", + "right": "issue:43986", + "accept": false, + "reason": "Qwen2.5-VL import failure and AutoProcessor video-model crash are different symptoms and likely different fixes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 45, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36683", + "issue:38617", + "issue:42490", + "issue:42757", + "issue:42907", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43208", + "issue:43299", + "issue:43352", + "issue:43502", + "issue:43526", + "issue:43550", + "issue:43582", + "issue:43618", + "issue:43756", + "issue:43761", + "issue:43792", + "issue:43824", + "issue:43825", + "issue:43867", + "issue:43906", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44162", + "issue:44222", + "issue:44230", + "issue:44368", + "issue:44380", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44655", + "issue:44683", + "issue:44938", + "issue:45290", + "issue:45478", + "issue:45479", + "issue:45698", + "issue:45753", + "issue:46123", + "issue:46132", + "issue:46172", + "issue:46178", + "issue:46209" + ], + "soft_pairs": [ + "issue:43066|issue:43906", + "issue:44938|issue:46178", + "issue:43208|issue:43550", + "issue:43992|issue:44484", + "issue:43825|issue:44655", + "issue:43012|issue:43550", + "issue:43012|issue:44368", + "issue:42907|issue:44655", + "issue:42907|issue:45753", + "issue:43352|issue:44380", + "issue:44162|issue:46172", + "issue:43352|issue:43976", + "issue:44380|issue:46132", + "issue:43792|issue:46123", + "issue:44380|issue:45479", + "issue:43064|issue:44380", + "issue:44483|issue:44514", + "issue:43208|issue:43526", + "issue:36683|issue:43582", + "issue:44222|issue:45478", + "issue:42757|issue:43824", + "issue:43012|issue:44683", + "issue:43618|issue:43761", + "issue:42490|issue:43012", + "issue:45698|issue:46172", + "issue:43618|issue:44380", + "issue:38617|issue:43502", + "issue:44038|issue:44230", + "issue:44483|issue:45290", + "issue:43756|issue:44485", + "issue:43066|issue:46209", + "issue:43299|issue:43867" + ], + "prepared_review_unit_hash": "4169caf30dee068a1611f1c85d7ff5ab6c8e94c561e2ec882881deabb462c5a2", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36942, + "estimated_input_tokens": 9236, + "estimated_eval_tokens": 18728 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a grab bag of unrelated Transformers issues; despite some superficial overlap (v5 regressions, model families, save/load paths), none of the soft-similarity pairs look like true duplicates or mergeable PR-equivalents.", + "confidence": 0.81, + "canonical_issue_reason": "issue:44483 is the most representative high-impact regression report in the set: a concrete v5 API break with clear user-facing impact and a specific endpoint involved.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44483 is the strongest standalone issue to keep: it is specific, severe, and actionable, with a clear failure mode that is easier to triage than the more model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Different bugs: tokenizer decoder type regression vs an isolated reproduction of another issue; no shared failing code path is evident." + }, + { + "left": "issue:44938", + "right": "issue:46178", + "accept": false, + "reason": "Python 3.14 load failure and a BACKENDS_MAPPING tensorflow_text backend error are unrelated import/config problems." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "xLSTM small-model training bugs and Bamba torch.compile SDPA failure affect different models and failure modes." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing embed_tokens.weight during UMT5 loading is unrelated to save_pretrained max_shard_size behavior." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "Pipeline task-support messaging regression and pipeline serialization failure are different code paths." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A bfloat16 precision warning during compile is not the same as a model-specific torch.compile SDPA crash." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "PyTorch float32 warning during compile and a tie_word_embeddings config warning are different issues." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Saving dequantized models and saving Pipeline objects involve different save_pretrained paths." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Dequantized save failure and a Qwen3.5 image infinite loop are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Unsupported Flash Attention 2 warning for Nemotron is unrelated to GPT2 attention scaling being ignored." + }, + { + "left": "issue:44162", + "right": "issue:46172", + "accept": false, + "reason": "ESM2 breakage and a DeepSeek tensor transpose bug are different model-specific defects." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "A model capability warning and a Python-version compatibility/import issue are not the same underlying bug." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "GPT2 backend scaling behavior and AttentionInterface.register side effects are distinct attention-stack changes." + }, + { + "left": "issue:43792", + "right": "issue:46123", + "accept": false, + "reason": "Whisper load/run failure and MaskGenerationPipeline dropping the final partial batch are unrelated pipeline/model bugs." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention scaling ignored in GPT2 is unrelated to a degenerate zero-loss classification bug." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Distributed optimizer-state corruption in Trainer/FSDP2 is not the same as GPT2 attention-scaling behavior." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Chat/completions request rejection and chat-template crashes on tool-call messages are different layers of the stack." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training blockers and BeitImageProcessorFast reduce_labels truncation are unrelated." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size and Apple Silicon allocator warmup TypeError are completely different failures." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "FP8 save_pretrained problems and Qwen3.5 MoE from_pretrained loading errors are separate issues." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Missing is_offline_mode from huggingface_hub and missing Qwen2_5_VLForConditionalGeneration from transformers are unrelated import errors." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "A warning during compilation and a hard flex_attention failure on torch>=2.9 are different backend problems." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "Both are CLIP regressions, but one is attentions missing and the other is hidden_states missing; they do not look like the same bug." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "A trailing-slash local model warning is unrelated to the bfloat16 compile warning." + }, + { + "left": "issue:45698", + "right": "issue:46172", + "accept": false, + "reason": "Wrong custom module restoration after save_pretrained and a DeepSeek dtype/transpose crash are different problems." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "Importing a removed symbol and local_files_only still making API calls are unrelated import vs network-behavior bugs." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Qwen3-VL-Moe loading regression and fp8 inference support are not the same concrete failure." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "Server request acceptance and apply_chat_template crashing on tool-call assistant messages are separate code paths." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE layer mismatch and GLM-5 RoPE implementation discussion do not describe the same defect." + }, + { + "left": "issue:43066", + "right": "issue:46209", + "accept": false, + "reason": "Tokenizer decoder type regression and FP8 inference failure on B200 are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Qwen3VL-Moe loading breakage and a state_dict sorting load error are different load-time 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": 46, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:36010", + "issue:36683", + "issue:41669", + "issue:42490", + "issue:42757", + "issue:42890", + "issue:42994", + "issue:43010", + "issue:43064", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43525", + "issue:43611", + "issue:43618", + "issue:43867", + "issue:43901", + "issue:43931", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44075", + "issue:44077", + "issue:44242", + "issue:44246", + "issue:44355", + "issue:44380", + "issue:44492", + "issue:44625", + "issue:44655", + "issue:44855", + "issue:44938", + "issue:45071", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45375", + "issue:45478", + "issue:45561", + "issue:45593", + "issue:45753", + "issue:45850", + "issue:46077", + "issue:46082", + "issue:46202" + ], + "soft_pairs": [ + "issue:36683|issue:43475", + "issue:45276|issue:46077", + "issue:42994|issue:44655", + "issue:43901|issue:44655", + "issue:36010|issue:42757", + "issue:43299|issue:45753", + "issue:44077|issue:44380", + "issue:36683|issue:44077", + "issue:43352|issue:43937", + "issue:43611|issue:44655", + "issue:44625|issue:46077", + "issue:43867|issue:45216", + "issue:46077|issue:46082", + "issue:43976|issue:44938", + "issue:43352|issue:45593", + "issue:36683|issue:43525", + "issue:43618|issue:45216", + "issue:36683|issue:44242", + "issue:43352|issue:43931", + "issue:43064|issue:44492", + "issue:43010|issue:45278", + "issue:43992|issue:45310", + "issue:43867|issue:46202", + "issue:44355|issue:44855", + "issue:43867|issue:45375", + "issue:41669|issue:44246", + "issue:36683|issue:43064", + "issue:44938|issue:45071", + "issue:42890|issue:45561", + "issue:43010|issue:45850", + "issue:43867|issue:45357", + "issue:42490|issue:44075", + "issue:43010|issue:43976", + "issue:43992|issue:45478", + "issue:43618|issue:44077" + ], + "prepared_review_unit_hash": "08859d77dc5680a8a0c12269b755af913541fb035eb9cf584d5f1a1c52e3fe61", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 35, + "serialized_chars": 37194, + "estimated_input_tokens": 9299, + "estimated_eval_tokens": 18854 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous cluster, but not only one plausible duplicate edge: 41669\u219444246 is a real import-slowdown duplicate candidate, and 45310\u219445478 are an exact-title Qwen3.5 from_pretrained pair that looks duplicate-like, with nearby Qwen3.5 save/load reports forming a loose family. Even so, the rest are largely separate model/config regressions, so the cluster still fits a broad umbrella issue rather than one tight bug.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45278 is still the broadest umbrella: it explicitly reports widespread import failures after a version upgrade, which best matches the cluster\u2019s general regression/import theme. It is broader than the Qwen3.5 mini-family, even though that family contains a duplicate-looking title pair.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best global representative because it captures the widest, clearest user-facing regression pattern in the set. The cluster also contains a smaller Qwen3.5 save/load family and an import-slowdown duplicate pair, but they are narrower than 45278\u2019s broad import-breakage report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Different models and different missing attributes (`vocab_size` vs `fpn_position_embeddings`); not the same bug." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Both are Gemma4-related, but one is token-embedding resizing and the other is attention-mask construction; separate code paths." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Both involve saving, but quantized-model saving and Pipeline `save_pretrained` are distinct failure modes." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "A docs mismatch and a pipeline serialization bug are not the same underlying issue." + }, + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different APIs/packages and different breakages." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Same model family, but one is a loading regression and the other is an image-specific generation loop." + }, + { + "left": "issue:44077", + "right": "issue:44380", + "accept": false, + "reason": "Unrelated model/config problems; no shared concrete code path." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Different model configs and different symptoms; only superficial similarity around optional config fields." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Flash Attention support gating and invalid `GenerationConfig` are unrelated failures." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "Loading with `base_model_prefix` and saving Pipeline objects are distinct serialization behaviors." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Different models and unrelated bugs (`num_labels` propagation vs attention-mask behavior)." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Both touch load/save correctness, but state-dict ordering during load and a Qwen3.5 save regression are not clearly the same bug." + }, + { + "left": "issue:46077", + "right": "issue:46082", + "accept": false, + "reason": "Different Gemma4/Llama config issues with unrelated validation and masking logic." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Both are Python-version compatibility complaints, but they target different versions and likely different root causes." + }, + { + "left": "issue:43352", + "right": "issue:45593", + "accept": false, + "reason": "Separate model features: Flash Attention support vs D-FINE auxiliary-loss handling." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Missing-config-attribute reports on different models and attributes; too generic to merge." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "Different regressions in different models/features; not mergeable into one fix." + }, + { + "left": "issue:36683", + "right": "issue:44242", + "accept": false, + "reason": "Unrelated model/config bugs; only both mention configuration behavior." + }, + { + "left": "issue:43352", + "right": "issue:43931", + "accept": false, + "reason": "Different model-loading problems with no shared concrete code path." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "Training/optimizer-state bug vs a typo in cache strategy names; no overlap beyond being in Transformers." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "A cache-update decoration fix and a broad import-regression report are unrelated." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "Both are `from_pretrained` errors, but for different models and different missing/malformed state." + }, + { + "left": "issue:43867", + "right": "issue:46202", + "accept": false, + "reason": "Sorted state_dict load failure and a Gemma pipeline roundtrip issue are different symptoms." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Both are import-time failures, but compiled Python files and a TorchScript parsing issue are unrelated triggers." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen3.5 checkpoint issues, but state-dict ordering and a missing vision config field are distinct bugs." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": true, + "reason": "Both describe slow `import transformers`; one points to wildcard imports as the likely cause and the other reports the same import-time slowdown symptom." + }, + { + "left": "issue:36683", + "right": "issue:43064", + "accept": false, + "reason": "Different failure domains: model config attribute missing vs distributed training optimizer-state corruption." + }, + { + "left": "issue:44938", + "right": "issue:45071", + "accept": false, + "reason": "Separate Python/runtime compatibility issues with different triggers and symptoms." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "Seed-related flakiness and an xdist file race are different testing bugs." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A specific cache/layer update fix and a vague general complaint about many models breaking are not the same bug." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Same broad `from_pretrained` theme, but different models and different reported missing/malformed state." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Both are Qwen3.5 checkpoint issues, but one is load failure from ordering and the other is incorrect visual encoder keys on save." + }, + { + "left": "issue:42490", + "right": "issue:44075", + "accept": false, + "reason": "Different issues: local-path import warnings vs optimizer SGD arguments being ignored." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache-update decoration and Python-version support are unrelated." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "Different regressions in different models/features; not mergeable into one fix." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The overall read is mostly conservative, but the summary overstates exclusivity: the packet also contains an exact-title duplicate-looking pair, issue:45310 and issue:45478, plus a small Qwen3.5 save/load family that could plausibly cluster together. Because of that, saying the import-time slowdown pair is the only plausible duplicate edge is not fully grounded in the packet." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 47, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:28282", + "issue:30333", + "issue:30990", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:36296", + "issue:42548", + "issue:42757", + "issue:42831", + "issue:42890", + "issue:43010", + "issue:43065", + "issue:43232", + "issue:43299", + "issue:43502", + "issue:43575", + "issue:43576", + "issue:43618", + "issue:43723", + "issue:43824", + "issue:43827", + "issue:43874", + "issue:44062", + "issue:44112", + "issue:44222", + "issue:44230", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44485", + "issue:44556", + "issue:44655", + "issue:44749", + "issue:44857", + "issue:44961", + "issue:44964", + "issue:44977", + "issue:45071", + "issue:45125", + "issue:45341", + "issue:45375", + "issue:45440", + "issue:45478", + "issue:45600", + "issue:46132" + ], + "soft_pairs": [ + "issue:36296|issue:43575", + "issue:28282|issue:33453", + "issue:43618|issue:44857", + "issue:44360|issue:45440", + "issue:43618|issue:44977", + "issue:42890|issue:44964", + "issue:43502|issue:44062", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:43723|issue:44556", + "issue:43576|issue:46132", + "issue:42548|issue:44351", + "issue:44387|issue:44556", + "issue:42890|issue:45341", + "issue:44222|issue:45125", + "issue:43010|issue:43065", + "issue:43502|issue:44485", + "issue:44355|issue:44749", + "issue:45125|issue:45478", + "issue:30333|issue:33453", + "issue:43010|issue:43232", + "issue:45125|issue:45375", + "issue:30990|issue:31795", + "issue:43502|issue:44222", + "issue:43299|issue:44556", + "issue:42831|issue:43874", + "issue:43827|issue:44961", + "issue:43502|issue:44961", + "issue:44355|issue:45071", + "issue:30333|issue:36296", + "issue:44315|issue:44360", + "issue:44355|issue:44655", + "issue:33453|issue:34689", + "issue:42757|issue:43576", + "issue:43576|issue:45071", + "issue:33453|issue:36296", + "issue:43576|issue:43824", + "issue:44222|issue:44230" + ], + "prepared_review_unit_hash": "f94c7a8d29b7c02c43e8e7fcdf48bd4ad4a3f40acf1eb7f98485202918003aca", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 38, + "serialized_chars": 38486, + "estimated_input_tokens": 9622, + "estimated_eval_tokens": 19500 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab bag of unrelated Transformers issues, with a loose theme around v5 loading/regression bugs and model-specific breakages. None of the soft pairs is clearly the same underlying defect, so I would keep them separate.", + "confidence": 0.34, + "canonical_issue_reason": "Issue 43723 is the cleanest representative of the cluster\u2019s common theme: a concrete, user-facing v5 loading regression with a direct API surface and broad relevance.", + "canonical_pr_reason": null, + "best_issue_reason": "43723 is the best anchor issue here because it is specific, actionable, and closest to the repeated \"loading broke in v5\" pattern seen across several items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Both mention tensor parallelism/loading, but one is a general tensor-parallel training bug and the other is an OOM while loading a specific model. Not the same concrete failure." + }, + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "PyTorch import error versus tokenizer loading regression. Same broad loader area, but different symptoms and code paths." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "CLIP output attentions bug and LwDETR AMP crash are unrelated model/runtime failures." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "DSA indexer/ReLU discussion is unrelated to DeepSeekV3MoE implementation divergence." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "Different models and failures: missing CLIP attentions versus Qwen3.5 flash-attention generation issues." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "Test seeding flakiness is unrelated to a Phi-4 multimodal loading regression." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "Local-files-only network leakage and AddedToken keyword collision are distinct bugs in different parts of the stack." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "One is a missing set_seed in integration tests; the other is a stale device-override CI failure. Not the same bug." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test seeding issue versus auto_mappings references to removed configs; unrelated failure modes." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Both are loading regressions, but one is tokenizer loading and the other is checkpoint reload after upgrade. Too broad to treat as one fix." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Broken env command and AttentionInterface.register behavior change are different features and failure paths." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different dependency/export issues." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Int4 CUDA memory regression and checkpoint reload failure are not the same underlying problem." + }, + { + "left": "issue:42890", + "right": "issue:45341", + "accept": false, + "reason": "Test utility bug is not the same as the SAM HQ seeding-related integration failure." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained bug and missing tensor-parallel plan for Qwen3.5 MoE are separate issues." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "Cache/layer update no_grad decoration and dummy Conv2d in Sam3PixelDecoder are unrelated." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "Network-request leakage on local_files_only and GLM-5 RoPE implementation are different bugs." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled Python file errors and a performance regression after upgrade are not the same issue." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Both involve Qwen3.5 MoE, but one is missing _tp_plan and the other is a broader from_pretrained error in v5.4.0. Not enough evidence of one concrete fix." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLFlow job reporting and tokenizer loading regression are unrelated." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "Both are generation-related, but one is cache/layer update decoration and the other is sync_gpus/_update_model_kwargs_for_generation behavior." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen3.5 MoE-family items, but one is tensor parallelism and the other is a missing vision config field. Different bugs." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers loading stall and documentation confusion around forward args are not duplicates." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "local_files_only network bug and FP8 save_pretrained moe bug are unrelated." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Different loading regressions on different model families and versions; not a single concrete defect." + }, + { + "left": "issue:42831", + "right": "issue:43874", + "accept": false, + "reason": "FP8 accuracy issue and missing image-processor method causing multimodal token counting failure are distinct." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Documentation references to removed pipeline APIs are unrelated to the placeholder 'racoon' issue." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "No relation between local_files_only networking and the 'racoon' issue." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file runtime errors and PretrainedConfig type checking regressions are different problems." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLFlow reporting failure and tensor-parallel training bug are unrelated." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel application with model_init and DSA indexer/ReLU discussion do not share a concrete bug." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Running compiled Python files and saving Pipeline objects are separate issues." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer loading regression and Llama 3.2 Vision model loading breakage are different model/load failures." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing is_offline_mode import from huggingface_hub is not the same as a broken transformers env command." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken env command and PretrainedConfig type-checking regression are unrelated." + }, + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Env command failure and Qwen2.5-VL import error are different issues." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "FP8 save_pretrained bug and FP8 inference support request are related only by topic, not by the same code-path 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": 48, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:31795", + "issue:36296", + "issue:38617", + "issue:41628", + "issue:41669", + "issue:42371", + "issue:42548", + "issue:42890", + "issue:43352", + "issue:43408", + "issue:43525", + "issue:43526", + "issue:43572", + "issue:43576", + "issue:43673", + "issue:43901", + "issue:43976", + "issue:44038", + "issue:44230", + "issue:44246", + "issue:44273", + "issue:44315", + "issue:44355", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44704", + "issue:44908", + "issue:44938", + "issue:44961", + "issue:45003", + "issue:45071", + "issue:45092", + "issue:45125", + "issue:45230", + "issue:45278", + "issue:45362", + "issue:45468", + "issue:45542", + "issue:45753", + "issue:45850", + "issue:46095", + "issue:46143", + "issue:46144", + "issue:46178", + "issue:46209" + ], + "soft_pairs": [ + "issue:41628|issue:42548", + "issue:31795|issue:38617", + "issue:44961|issue:45468", + "issue:44355|issue:46178", + "issue:43576|issue:44938", + "issue:44315|issue:46209", + "issue:46143|issue:46144", + "issue:44908|issue:44961", + "issue:44485|issue:44961", + "issue:36296|issue:43901", + "issue:43673|issue:44492", + "issue:36296|issue:43526", + "issue:45362|issue:45850", + "issue:44623|issue:45230", + "issue:44273|issue:45542", + "issue:44038|issue:45125", + "issue:44273|issue:45003", + "issue:44230|issue:46095", + "issue:44393|issue:45071", + "issue:43576|issue:44556", + "issue:43525|issue:45850", + "issue:44393|issue:45753", + "issue:41669|issue:43673", + "issue:44556|issue:45092", + "issue:43572|issue:45850", + "issue:42371|issue:44492", + "issue:44492|issue:45850", + "issue:45278|issue:45850", + "issue:44246|issue:44273", + "issue:42890|issue:44908", + "issue:44704|issue:44908", + "issue:43408|issue:44315", + "issue:43976|issue:44393", + "issue:43352|issue:44393" + ], + "prepared_review_unit_hash": "4ed796aeebfa94850c87e4b03641e5694089ed24a0b05e05bd4539f3d5129604", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37146, + "estimated_input_tokens": 9287, + "estimated_eval_tokens": 18830 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mix of unrelated model/feature bugs, with one dominant subtheme around v5 import/export regressions and lazy-loading/import behavior. The clearest representative issue is the broad import-regression report.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45278 best captures the main cluster theme: widespread import failures after upgrading to v5.5.0, which is the most broadly representative and actionable report here.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the strongest umbrella issue for the import-error/lazy-loading regression subset; the other items are either narrower, model-specific, or unrelated.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import errors, but they target different symbols and there is no evidence they are the same concrete missing-export bug." + }, + { + "left": "issue:31795", + "right": "issue:38617", + "accept": false, + "reason": "Documentation confusion about forward args is unrelated to an ImportError for layer_type_validation." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "These are completely unrelated issues; one is a placeholder title and the other is a Gemma audio positional-encoding bug." + }, + { + "left": "issue:44355", + "right": "issue:46178", + "accept": false, + "reason": "Compiled Python file failures and a tensorflow_text backend mapping error are different failure modes with different code paths." + }, + { + "left": "issue:43576", + "right": "issue:44938", + "accept": false, + "reason": "A broken env command and a Python 3.14 load failure are distinct issues, not the same regression." + }, + { + "left": "issue:44315", + "right": "issue:46209", + "accept": false, + "reason": "Liger Kernel application in model_init and FP8 inference on B200 are separate model/runtime problems." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "Both touch RoFormer, but one is missing kwargs propagation while the other is about using the attention interface; not the same bug." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs handling is unrelated to the unrelated placeholder issue." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM-5 RoPE implementation discussion is unrelated to the placeholder issue." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor-parallel training bugs and a documentation mismatch for return_all_scores are not the same underlying problem." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "A cache-missing runtime bug and a typo in cache strategy docs/code are different issues." + }, + { + "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": "A specific Qwen3.5-35B crash is not the same as a vague report that many models break in v5." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Processor save_pretrained missing files is a concrete bug; the other is a generic, underspecified report." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading correctness and an undefined tf backend error from tensorboard-only installs are different bugs." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL-Moe support and missing _tp_plan for tensor parallelism are related to the model family but not the same fixable code-path issue." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading failures and unsafe sys.modules access are adjacent areas, but the reported bugs are distinct." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "FP8 inference support for Qwen3-VL and a weight-conversion question for DeepSeekV4 are unrelated." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "A Qwen3-VL bounding-box output bug and a PretrainedConfig type-checking break are different problems." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Broken env command behavior and checkpoint reload incompatibility across versions are separate regressions." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "A missing pad_token_id on Llama4Config is a specific config bug, not the same as a broad complaint that many models break." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig is a specific config bug, not the same as the generic v5 complaint." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A cache-strategy typo is not the same underlying bug as a broad model-breakage report." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Both are about v5 regressions, but 45278 is a concrete import-error report while 45850 is too vague to be the same specific bug." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Import slowdown and broken lazy loading are different symptoms and likely different fixes." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Missing set_seed in tests and an inverse_sqrt scheduler kwargs bug are unrelated." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor cached_file kwargs handling and scheduler kwargs handling are unrelated code paths." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "A wrong model-type warning and a Liger Kernel/model_init application bug are not the same issue." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python-version compatibility and a Qwen3-VL output bug are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2 support for NemotronH and Qwen3-VL bounding-box errors are different model-specific 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": 49, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:41628", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:43066", + "issue:43298", + "issue:43502", + "issue:43519", + "issue:43576", + "issue:43606", + "issue:43618", + "issue:43704", + "issue:43827", + "issue:43976", + "issue:44077", + "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:45230", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45850", + "issue:46018", + "issue:46095", + "issue:46172" + ], + "soft_pairs": [ + "issue:44297|issue:45230", + "issue:44295|issue:45230", + "issue:43576|issue:44908", + "issue:41669|issue:44351", + "issue:44908|issue:45071", + "issue:41669|issue:45341", + "issue:44492|issue:44945", + "issue:43976|issue:45850", + "issue:44393|issue:45478", + "issue:44077|issue:44861", + "issue:43827|issue:44908", + "issue:41628|issue:41669", + "issue:31795|issue:33453", + "issue:44315|issue:44829", + "issue:43576|issue:44273", + "issue:44655|issue:44861", + "issue:44360|issue:45468", + "issue:45230|issue:45310", + "issue:46018|issue:46172", + "issue:44273|issue:44908", + "issue:43519|issue:44485", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:31795|issue:34689", + "issue:31515|issue:31795", + "issue:43704|issue:44485", + "issue:44485|issue:45468", + "issue:43298|issue:46095", + "issue:42371|issue:43704", + "issue:43502|issue:43519", + "issue:43519|issue:44623", + "issue:43519|issue:43704", + "issue:43606|issue:45468", + "issue:44263|issue:45468", + "issue:44261|issue:45468", + "issue:42913|issue:43066", + "issue:44908|issue:44998", + "issue:44492|issue:44998", + "issue:43618|issue:44998", + "issue:44861|issue:45290", + "issue:44998|issue:45850", + "issue:44998|issue:45341" + ], + "prepared_review_unit_hash": "172b2082a6f6ae6dd14e3314af71ed7601d991bf6d369b4a456dcfe0c1984138", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 39485, + "estimated_input_tokens": 9872, + "estimated_eval_tokens": 20000 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: most soft edges connect clearly different bugs, docs issues, or unrelated subsystems. I would keep the earliest concrete Qwen3.5 MoE from_pretrained regression as the representative issue, but I would not merge any of the soft-edge pairs.", + "confidence": 0.86, + "canonical_issue_reason": "issue:45310 is the earliest concrete report of the repeated Qwen3.5 MoE from_pretrained failure and is specific enough to represent that regression thread better than the broader or unrelated items in the cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the best issue to keep as the representative record: it has a concrete model-loading failure, a clear version trigger, and enough detail to be actionable.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "A tokenizer save_pretrained metadata mismatch is not the same as a generic placeholder bug report." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Reading a buffer-registered position_ids tensor is a distinct bug from a generic placeholder report." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "A broken env CLI command and a scheduler-kwargs regression are unrelated code paths." + }, + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Removing import * for performance is unrelated to a missing HybridCache import." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs handling and PretrainedConfig type checking are different problems." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "A model import cleanup/performance issue is unrelated to a testing_utils bug." + }, + { + "left": "issue:44492", + "right": "issue:44945", + "accept": false, + "reason": "A cache-strategy typo is not the same as incorrect pipeline-parallelism output." + }, + { + "left": "issue:43976", + "right": "issue:45850", + "accept": false, + "reason": "Python 3.9 support breakage and a broad 'too many models break' complaint are not the same bug." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Different models and failure modes: Qwen3-VL 2D box output versus Qwen3.5 MoE from_pretrained loading." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "Optional post_init policy for patchtsmixer is unrelated to a tied-weights AttributeError." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs still referencing pipeline() after removals is unrelated to inverse_sqrt scheduler kwargs." + }, + { + "left": "issue:41628", + "right": "issue:41669", + "accept": false, + "reason": "AutoImageProcessor import failures and import * cleanup/performance are different import issues." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Forward docs confusion is not the same as a tokenizer loading regression." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger Kernel not applied with model_init and flash_attention_3 training degeneration are different acceleration/training bugs." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Broken env command and lazy loading failure are unrelated." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "Saving Pipeline objects and tied-weight key extraction are different serialization paths." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "A DSA indexer activation issue is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "A generic placeholder report is not the same as a concrete Qwen3.5 MoE loading regression." + }, + { + "left": "issue:46018", + "right": "issue:46172", + "accept": false, + "reason": "DeepSeek shared-expert gating and a comb transpose bug are different model implementation issues." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading is unrelated to scheduler argument propagation." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Qwen3VL timestamp calculation and GLM-5 RoPE implementation are different numeric/model bugs." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Timestamp computation in a processor is unrelated to scheduler kwargs." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import cleanup/performance is unrelated to a Qwen3 VRAM leak." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion about forward arguments is unrelated to Llama 3.2 model loading failure." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "Slow checkpoint loading and forward-doc confusion are different issues." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "A VRAM leak and a RoPE implementation discussion are different bugs." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "GLM-5 RoPE behavior and Gemma-4 audio positional encoding are unrelated model-specific math paths." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Audio continuous batching support and a DeepSeek-V4 weight-conversion question are not the same bug." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 API behavior and a Qwen3 VRAM leak are unrelated." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "Unexpected API requests under local_files_only and Qwen3VL timestamp math are unrelated." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Processor timestamp calculation and processor.save_pretrained missing files are different failures." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Processor timestamp math and a multi-thread VRAM leak are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch and Gemma-4 audio positional encoding are different problems." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "A torch.split return-value issue is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "Missing rms_norm_eps in MLA q_a_layernorm is unrelated to Gemma-4 audio positional encoding." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "Both mention tokenizer v5 changes, but one is a broad behavior difference and the other is a specific decoder-type bug; not the same underlying defect." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "A scheduler kwargs regression is unrelated to a nonsense placeholder issue." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "A cache-strategy typo is unrelated to a nonsense placeholder issue." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression is unrelated to a nonsense placeholder issue." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weight key extraction and chat template tool-call crashes are different code paths." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "A nonsense title is not the same as a broad model-breakage complaint." + }, + { + "left": "issue:44998", + "right": "issue:45341", + "accept": false, + "reason": "A nonsense placeholder issue is unrelated to a testing_utils 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": 50, + "review_unit_count": 50, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-430", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:33453", + "issue:36296", + "issue:37934", + "issue:38617", + "issue:42936", + "issue:43240", + "issue:43298", + "issue:43525", + "issue:43572", + "issue:43668", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:44230", + "issue:44623", + "issue:44861", + "issue:45310", + "issue:46018", + "issue:46144" + ], + "soft_pairs": [ + "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:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230", + "issue:43717|issue:46144", + "issue:42936|issue:46018", + "issue:43668|issue:44031", + "issue:42936|issue:43240", + "issue:43240|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:46144", + "issue:42936|issue:43813", + "issue:43240|issue:43668", + "issue:43668|issue:43813", + "issue:37934|issue:43668", + "issue:42936|issue:43668" + ], + "prepared_review_unit_hash": "2aa9b4d7065669e5936d6e4e47353ceac550b4cd5eae2f8325523f8cc36ea52c", + "component_budget": { + "node_count": 430, + "item_count": 430, + "soft_pair_count": 1591, + "serialized_chars": 823302, + "estimated_input_tokens": 205826, + "estimated_eval_tokens": 411908 + }, + "budget": { + "node_count": 21, + "item_count": 21, + "soft_pair_count": 21, + "serialized_chars": 18006, + "estimated_input_tokens": 4502, + "estimated_eval_tokens": 9260 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is over-broad and mostly driven by title similarity: it mixes unrelated issues across loading, tokenizers, configs, model internals, training, and inference. None of the suggested soft pairs look like the same underlying bug.", + "confidence": 0.93, + "canonical_issue_reason": "No single canonical issue is a good fit here\u2014the items are too heterogeneous, and the soft edges are only lexical similarity matches rather than true duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "If forced to pick one representative, issue 43240 is the broadest concrete runtime-bug report in the set (loss kwargs propagation) and has the most inbound references, but it still does not duplicate the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Different bugs: one is missing pad_token_id on Llama4Config, the other is an AttributeError in tied-weight key handling." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "StableLm pad_token_idx regression and _tied_weights_keys list crash are separate config/runtime issues." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Checkpoint loading slowness and a missing import from configuration_utils are unrelated code paths." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 inference support for Qwen models is a different change from processor.save_pretrained missing files." + }, + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "MLflow job status reporting on failed training and slow from_pretrained checkpoint loading are unrelated." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "One is loading performance, the other is a tensor-parallel training bug; not the same defect." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "FP8 support request and Qwen3.5 MoE from_pretrained failure are different issues, not one duplicate bug." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Both involve loading, but one is slow checkpoint loading while the other is a tokenizer-loading regression." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching for audio models and Qwen3-VL/Qwen3.5 FP8 support are different feature gaps." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba/Mamba-2 init_weights behavior and RoFormer attention-interface usage are unrelated architecture-specific issues." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Mask2Former ignore_value handling and DeepSeek-V4 shared-expert gating are different model bugs." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "ModernBERT type-hint correctness and tokenizer regex warning behavior are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Both touch training/loss plumbing, but ignore_value not used and kwargs not passed to loss are different concrete bugs." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Loss-kwargs forwarding and RoFormer attention-interface usage do not describe the same code-path problem." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Mask2Former ignore_value usage and tokenizer regex warnings are unrelated subsystems." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "Mask2Former loss handling and RoFormer attention implementation are not duplicates." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "A model loss-parameter bug and a quantization typo are clearly different issues." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Loss-function kwargs forwarding and an incorrect type hint are not the same bug." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Incorrect type annotation and a typo in a quantization attribute path are unrelated." + }, + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "A question about RMS norm behavior and a ModernBERT type-hint issue are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2Former ignore_value handling and ModernBERT norm_eps typing are different 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": 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-10519-3", + "nodes": [ + "issue:10519", + "issue:11005", + "issue:11307" + ], + "soft_pairs": [ + "issue:10519|issue:11307", + "issue:10519|issue:11005", + "issue:11005|issue:11307" + ], + "prepared_review_unit_hash": "3db553639ca2a85c89f00059655b735feb34b5b8dc99aa6e23e2f443dfb6000a", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2876, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 2876, + "estimated_input_tokens": 719, + "estimated_eval_tokens": 1694 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The three issues are only superficially similar in wording and appear to be unrelated feature requests: truncation behavior, ReduceLROnPlateau-like scheduling, and Wav2Vec2 word time offsets. None look like duplicates of the same underlying bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue 10519 is the most concrete and self-contained request, with a clearly specified behavior change and scope.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 10519 is the best representative of the cluster only because it is the most specific and actionable; the other two are unrelated topics.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:10519", + "right": "issue:11307", + "accept": false, + "reason": "Different domains and requested changes: token truncation behavior vs. word boundary timestamps in Wav2Vec2." + }, + { + "left": "issue:10519", + "right": "issue:11005", + "accept": false, + "reason": "One is about truncation direction, the other about learning-rate scheduling functionality; no shared underlying bug or fix." + }, + { + "left": "issue:11005", + "right": "issue:11307", + "accept": false, + "reason": "Unrelated feature requests in different subsystems, with no concrete overlap in code path or expected change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 3, + "component_count": 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": "c1b07b3e511c524eeed707235846b5c44d606a601726a6325f18536c6bc6201e", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2477, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2477, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These issues do not look like duplicates of one another. #16618 is a model-loading failure report, while #15980 is about a bad login/private-model download error message and #13244 is about Tapas tokenization differing from TensorFlow code.", + "confidence": 0.87, + "canonical_issue_reason": "#16618 is the most concrete and broadly representative of the cluster because it is the only issue directly about a failed model load, making it the best anchor for the set even though the other issues are not true duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the best single issue to represent the cluster because it is the most central user-facing failure case among the three; however, it should not be treated as a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tapas tokenization mismatch and inability to load bert-base-uncased are different problems affecting different code paths; they do not describe the same underlying bug." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "A private-model download/authentication error message is unrelated to a generic 'can't load bert-base-uncased' failure; same subsystem perhaps, but not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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-26350-2", + "nodes": [ + "issue:26350", + "issue:33260" + ], + "soft_pairs": [ + "issue:26350|issue:33260" + ], + "prepared_review_unit_hash": "740361ac90c2811f1e0640057570bf7ae02b8fdafde926bbdd0a292547853382", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1590, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are both community contribution feature requests for expanding model/runtime support, but they target different capabilities: Flash Attention 2 vs GGUF. They do not look like the same underlying bug or change, so they should not be deduplicated.", + "confidence": 0.98, + "canonical_issue_reason": "If a representative issue is needed, #33260 is the better anchor because it is newer and has inbound references, but it is not a duplicate of #26350.", + "canonical_pr_reason": null, + "best_issue_reason": "#33260 is the better cluster representative only in a loose organizational sense; it is more recent and externally referenced, though it covers a different feature area than #26350.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different support requests with different concrete targets: Flash Attention 2 architecture coverage vs GGUF architecture coverage. Same broad theme, but 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": "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-27088-6", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:38435", + "issue:43834", + "issue:44365", + "issue:44370" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:38435", + "issue:27088|issue:28936", + "issue:28936|issue:43834", + "issue:28936|issue:44365", + "issue:28936|issue:38435", + "issue:28936|issue:44370", + "issue:38435|issue:44370", + "issue:38435|issue:44365", + "issue:38435|issue:43834", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:44365|issue:44370" + ], + "prepared_review_unit_hash": "988b9a1040506a9b172fc1ba513f55db238f1c38341126cdde23c3fcbb6718cd", + "component_budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 15, + "serialized_chars": 8789, + "estimated_input_tokens": 2198, + "estimated_eval_tokens": 4652 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 15, + "serialized_chars": 8789, + "estimated_input_tokens": 2198, + "estimated_eval_tokens": 4652 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are separate i18n translation-request issues, not duplicates of the same underlying bug/change. The Turkish, Spanish, and Romanian tickets are language-specific, and the generic placeholder-titled issues still read like distinct translation requests rather than one shared fix.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Both are translation requests, but there is no evidence they target the same language or the same concrete work item." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Turkish docs translation is a distinct request from the other issue; same broad category is not enough to merge." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Different translation ticket; the placeholder title does not show this is the same language or task as the Turkish issue." + }, + { + "left": "issue:27088", + "right": "issue:38435", + "accept": false, + "reason": "Turkish and Romanian docs translations are separate language-specific issues, not duplicates." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Turkish and Spanish docs translations are different requests for different target languages." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish docs translation is distinct from a generic placeholder issue; no concrete shared change is demonstrated." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "These are both translation-related, but the available evidence does not show the same target language or same underlying task." + }, + { + "left": "issue:28936", + "right": "issue:38435", + "accept": false, + "reason": "Spanish and Romanian translation requests are separate issues." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "No evidence they are the same translation ticket; only the broad docs-i18n theme overlaps." + }, + { + "left": "issue:38435", + "right": "issue:44370", + "accept": false, + "reason": "Romanian translation is distinct from the placeholder-titled issue; not enough to treat as the same request." + }, + { + "left": "issue:38435", + "right": "issue:44365", + "accept": false, + "reason": "Romanian docs translation and the placeholder issue are not shown to be the same language-specific task." + }, + { + "left": "issue:38435", + "right": "issue:43834", + "accept": false, + "reason": "Separate translation requests; shared subsystem is too broad to merge." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "The two placeholder issues may be similar in wording, but there is no concrete evidence they describe the same language or work item." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "Likely template-like translation issues, but the data does not establish the same underlying request." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "Both look like generic translation tickets, but not enough detail to conclude they are the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 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-36817-16", + "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", + "issue:46133" + ], + "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", + "issue:45322|issue:46133" + ], + "prepared_review_unit_hash": "64555a2a4072fde57243d5f920b2bf87394e0d0f7b6008d735827e3f9900fe95", + "component_budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 36, + "serialized_chars": 21657, + "estimated_input_tokens": 5415, + "estimated_eval_tokens": 11086 + }, + "budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 36, + "serialized_chars": 21657, + "estimated_input_tokens": 5415, + "estimated_eval_tokens": 11086 + }, + "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 that share a generic \"add model support\" pattern, not the same underlying bug/change. No PRs are present.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43671 is the most central open request in this set by similarity fan-in, so it is the best single representative of the cluster's broad model-request theme.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43671 is the best global issue representative: open, broadly connected to other request-like issues, and close to the center of the similarity graph, though still not a duplicate of them.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different requested models (EuroBert vs DEIMv2); only the generic \"add model\" framing overlaps." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "EuroBert and Qwen3-TTS are unrelated model additions." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "EuroBert vs RF-DETR are different architectures and use cases." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "RF-DETR and DEIMv2 are separate model requests, not the same change." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "Object detection model request vs audio model request; unrelated." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "Different model families and code paths; only broad request similarity." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "RF-DETR and Qwen3-ASR are unrelated model additions." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "RF-DETR vs Qwen3-TTS are different model support requests." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "NeoBERT and Qwen3-ASR are unrelated; same generic request theme only." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "NeoBERT vs weight-sparse transformer are different concrete model changes." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "BERT model request vs audio model request; not the same underlying change." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "RF-DETR and NeoBERT are different models and code paths." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "NeoBERT and ModernVBERT are different model requests despite similar naming." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "MiniCPM3 architecture vs Qwen3-ASR are distinct model integrations." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "MiniCPM3 and the weight-sparse transformer request are unrelated." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "MiniCPM3 architecture and ModernVBERT are separate model additions." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "MiniCPM3 and Voxtral TTS are different model families and tasks." + }, + { + "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 model support requests." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "ModernVBERT and Qwen3-TTS are unrelated." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "ModernVBERT and generic BERT-like RoPE support are not the same concrete change." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "Generic RoPE/BERT-variant request vs Voxtral TTS support; different targets." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Same vendor family, but TTS and ASR are distinct models/features and would not be one merged PR." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Both are audio-related, but Qwen3-TTS and Voxtral TTS are separate model integrations." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated model requests." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "TTS support and a weight-sparse transformer proposal are different code-path changes." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "Qwen3-TTS and LFM2.5 Audio are separate model additions." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "ASR and LFM2.5 Audio are different concrete model support requests." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Qwen3-ASR and EUPE are unrelated model architectures/tasks." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "ASR support and weight-sparse transformer support are different changes." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "Audio model request vs SAM3-LiteText; no shared underlying bug/change." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Weight-sparse transformer and EUPE are distinct proposals." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Qwen3-ASR and Voxtral TTS are separate audio model integrations." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "NeoBERT and Voxtral TTS are unrelated model requests." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "ASR support and OCR2 support are different model additions." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "Both are vision-language related, but they target different architectures and objectives, not one duplicate change." + } + ] + }, + "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": "21c10e9a8e8dccf2a940d514e183b60a4a95424da65636112da8158521c2e6ed", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1480, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are not duplicates: one is a general request about block decoding, while the other asks for DeepSeek v3.2 support. They concern different features with no shared targets or code path evidence.", + "confidence": 0.96, + "canonical_issue_reason": "No canonical issue; the requests are unrelated and do not describe the same bug or feature.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43037 is the more specific, actionable request, but it is not a duplicate of 41084.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different topics: block decoding behavior vs model support request; no evidence they describe 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": 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": "0703ab5d91e0f87bf6b6373538982b9e807704471d679825296fb93497b57fe2", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1584, + "estimated_input_tokens": 396, + "estimated_eval_tokens": 1048 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related only at a very high level (both concern Trainer custom behavior), but they target different code paths: one is about `remove_unused_column` and input-column handling, the other is about FLOPs estimation for custom trainers. They should not be treated as duplicates.", + "confidence": 0.92, + "canonical_issue_reason": "issue:43598 is the cleaner representative because it describes a concrete Trainer customization problem around `remove_unused_column`, with a more specific code-path focus than the FLOPs-estimation request.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43598 is the better standalone issue to keep as the cluster representative since it is narrowly scoped and likely easier to act on; issue:43600 addresses a separate instrumentation/metrics concern.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Reject: the titles and implied code paths differ. `remove_unused_column` affects data-column pruning/custom trainer behavior, while FLOPs estimation is a separate accounting/metrics feature. They are not the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "cd614fae16889f9430e48c54c39c0f50440e604419ff611a08cfe166acaa4890", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These two issues are both about Qwen model support, but they target different gaps: one requests registration for Qwen3-Omni in generic auto classes, while the other asks for sequence-classification support for Qwen3.5. They are related by ecosystem area, not the same underlying bug or change.", + "confidence": 0.9, + "canonical_issue_reason": "Issue #43684 is the broader and more foundational request: adding Qwen3-Omni registration to AutoModel / AutoModelForConditionalGeneration. It is a direct model-registration gap rather than a single task-specific head.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #43684 is the better representative of the cluster because it covers the core auto-class registration problem, whereas #44405 is a separate, narrower feature request for AutoModelForSequenceClassification on a different config.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different targets and likely different fixes: generic Qwen3-Omni auto-model registration versus sequence-classification support for Qwen3.5. They share a model family theme, but not the same concrete change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "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": "57e3cef988e4209a66fcb9828c80beb0bf14420181f5a73922e5687d9e861440", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4263, + "estimated_input_tokens": 1066, + "estimated_eval_tokens": 2388 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "43845, 43846, and 43859 look like a tentative same-title cluster, but the packet only shows title similarity, close timestamps, and soft-similarity scores; that is not enough to confidently label them the same duplicate issue. 43835 looks unrelated. No PRs are present.", + "confidence": 0.54, + "canonical_issue_reason": "issue:43845 is the best available representative for the similar 'huggingface' issues because it is the earliest of that small title-matched set, but the duplicate relationship is only tentative from the available metadata.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43845 is the best overall issue candidate because it is the earliest member of the tentatively related 'huggingface' group and there is no stronger distinguishing evidence for the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "Different titles and no body/link evidence tying them to the same underlying report; the metadata does not support a duplicate match." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "Different titles and no supporting evidence of a shared bug/report." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "Different titles and no supporting evidence of a shared bug/report." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": false, + "reason": "Same short title and close timing suggest possible relation, but title similarity alone is not enough to confirm the same underlying issue." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "The packet lacks body content or explicit linkage; identical titles and timing make them plausible duplicates, but not confirmable." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "Only title similarity and close timestamps are available, which is insufficient to accept this as the same issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The analysis is a bit too strong for the packet. The only evidence for 43845/43846/43859 is identical titles plus close timestamps and soft-similarity scores; the packet does not provide body content or explicit linkage, and all soft pairs have deterministic_accept=false. It is reasonable to say these three are likely related or may be duplicate reports, but not to state confidently that they are the same duplicate issue. The rejection of 43835 as unrelated is fine. Please soften the cluster/canonical wording and avoid overclaiming duplicate identity from title similarity alone." + }, + "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": "fa76c0604c8faba6242518d32a002cc89de4080776a94ca87a75748c0a47fdfe", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5788, + "estimated_input_tokens": 1447, + "estimated_eval_tokens": 3150 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5788, + "estimated_input_tokens": 1447, + "estimated_eval_tokens": 3150 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster looks like a mix of spam/placeholder issues rather than one real duplicated bug report. The three \"\" issues are similar in title only, while the documentation and \"Google LLC\" items are unrelated low-signal posts.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44371 is the earliest of the repeated \"\" entries, so it is the least-bad anchor if a canonical node is required, but the cluster does not show a real shared issue.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44418 is the only item with a non-placeholder title and is the most readable representative, though it still does not look like a true duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are just \"\" placeholders with no evidence of the same underlying bug or request beyond identical noise." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "Identical spam-like titles are not enough to treat them as the same issue; there is no substantive content linking them." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "One is a spam placeholder and the other is a generic documentation suggestion; they do not describe the same problem." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "A spam placeholder and a \"Google LLC\" promo-style post are unrelated." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "Generic documentation suggestion versus spam placeholder; no shared underlying request is visible." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "Generic documentation suggestion versus spam placeholder; not the same issue." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Same placeholder title, but there is no concrete bug/change described to justify duplicate merging." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "Two different spam-like posts with no evidence of a shared topic or fix." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "No substantive overlap beyond low-quality content; these are not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "8c3d91dbf076e1d3e4c5c576e91b3e037d1feaa724defdbdfb0f7588f2a93a97", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1646, + "estimated_input_tokens": 412, + "estimated_eval_tokens": 1080 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two closed issues are likely related and may be duplicates, but the evidence is not strong enough to call them a certain duplicate. They share an identical title and moderate soft similarity, while the packet does not show near-identical bodies or stronger deterministic signals. #45145 remains the better canonical representative because it has the same topic plus the only inbound reference.", + "confidence": 0.84, + "canonical_issue_reason": "#45145 is the best canonical issue because it matches the same title/topic as #45151 and has slightly stronger linkage via one inbound reference.", + "canonical_pr_reason": null, + "best_issue_reason": "#45145 is a reasonable cluster representative for this issue pair because it appears to be the more referenced copy of the same topic.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "The pair has identical titles and a moderate soft-similarity score, but the packet does not provide enough evidence to conclude a clear duplicate with high confidence. The conservative read is that they are probably related, not definitively the same issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The canonical choice is reasonable, but the duplicate verdict is too strong for the evidence provided. The packet only shows identical titles and a soft_similarity signal (jaccard ~0.63) with deterministic_accept=false; it does not establish \u201cnear-identical\u201d bodies or a \u201cclear duplicate\u201d with high confidence. Please soften the wording and keep the conclusion conservative." + }, + "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": "384d10d13b8de50fa77712737a482ee7f4d8bb8632513a6354365435e9a45ae2", + "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 are two separate 4-bit inference bugs in the same broad area, but they do not look like duplicates: one is a Gemma4 audio-module uint8/torch.finfo TypeError, the other is a BitsAndBytesConfig skip_modules regression that drops the default lm_head exclusion and triggers an AssertionError.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Different failure modes and code paths: 45672 is Gemma4/audio-module dtype handling during NF4 inference, while 45674 is a config regression around llm_int8_skip_modules clearing the default lm_head exclusion. Similar 4-bit context, but not the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "46b0a9a374a692058f62728ea176bd577ccdf2d8af148f95419c3c6bfe27b37e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1476, + "estimated_input_tokens": 369, + "estimated_eval_tokens": 994 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1476, + "estimated_input_tokens": 369, + "estimated_eval_tokens": 994 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both issues appear to be unrelated spam/personal-name entries with no technical content; they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": "Issue 45762 is marginally earlier, but neither issue has substantive technical detail to serve as a meaningful canonical report.", + "canonical_pr_reason": null, + "best_issue_reason": "If one must be retained as a representative, 45762 is slightly better only because it was created first; otherwise neither is suitable as a canonical issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "The titles are different personal names with different dates and there is no shared bug, feature, or code-path context; this is not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 16, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45764-2", + "nodes": [ + "issue:45764", + "issue:45765" + ], + "soft_pairs": [ + "issue:45764|issue:45765" + ], + "prepared_review_unit_hash": "68e86a5578e7ed3f85cac345652c5d012325157b2154cb54fd8c996f86bdafc6", + "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": "Both items are closed issues with only personal-name/date titles and no body or technical context; they do not appear to describe the same bug or change. This looks like unrelated spam/miscellaneous entries rather than a duplicate cluster.", + "confidence": 0.98, + "canonical_issue_reason": "Neither issue has enough technical content to serve as a canonical bug report; they are just name/date titles and appear unrelated.", + "canonical_pr_reason": null, + "best_issue_reason": "No issue in this pair is a good canonical candidate because there is no shared underlying problem to preserve.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "The titles refer to different personal names and dates, with no description, links, or evidence of a shared bug/change; not a duplicate." + } + ] + }, + "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": "a720d5be17ae9edb80637cf8fdb696adba8ff2244c3724003a6318daa5c2971d", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5722, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5722, + "estimated_input_tokens": 1431, + "estimated_eval_tokens": 3118 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is not one single duplicate set: it contains two exact duplicate issue pairs (45870/45871 and 45872/45873) plus one separate, related-but-distinct issue (45869).", + "confidence": 0.93, + "canonical_issue_reason": "Issue 45870 is the best canonical issue for its pair: it is an exact duplicate of 45871, appears slightly earlier, and has the clearest self-contained title. Note that 45872/45873 form a separate duplicate pair, not the same underlying request.", + "canonical_pr_reason": null, + "best_issue_reason": "45870 is the best issue representative overall because it is an exact duplicate pair leader and the earliest clear duplicate in the set, but it should not absorb the unrelated 45869 or the separate 45872/45873 request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Both mention financial data/time series, but 45869 is a dataset bundle offer and 45870 is an example request; different underlying issue." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Same as 45869 vs 45870: related topic, but a dataset offer is not the same as an example request." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Exact same title and very likely same body/content; these are clear duplicates of the same issue." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Both are DataForge financial/time-series items, but one is an example request and the other includes a scraping service; different requests." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Same as 45870 vs 45872; similar theme, but not the same concrete issue." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Despite shared keywords, these describe different asks: example request vs time-series+scraping service." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Same as 45871 vs 45872; not the same underlying request." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Exact same title and matching content pattern; clear duplicate issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: 45870/45871 and 45872/45873 are clearly supported as duplicate pairs by identical titles and matching bodies, while 45869 is only loosely related and should stay separate. The soft-edge verdicts are conservative and do not overstate evidence." + }, + "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-39895-13", + "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:46048" + ], + "soft_pairs": [ + "pull_request:39895|pull_request:43424", + "pull_request:41855|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:42668|pull_request:43823", + "pull_request:43251|pull_request:43823", + "pull_request:43251|pull_request:44827", + "pull_request:43424|pull_request:44827", + "pull_request:43424|pull_request:43823", + "pull_request:43823|pull_request:45655", + "pull_request:43823|pull_request:43995", + "pull_request:43995|pull_request:45213", + "pull_request:43995|pull_request:44827", + "pull_request:44827|pull_request:45170", + "pull_request:44827|pull_request:46048", + "pull_request:44827|pull_request:45655", + "pull_request:44827|pull_request:45213", + "pull_request:43823|pull_request:45213", + "pull_request:43251|pull_request:43254" + ], + "prepared_review_unit_hash": "9c82aec2a7de057e1023f03e9842db4134d385deffb6a782d7843aabf89e610c", + "component_budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 18, + "serialized_chars": 34769, + "estimated_input_tokens": 8693, + "estimated_eval_tokens": 17642 + }, + "budget": { + "node_count": 13, + "item_count": 13, + "soft_pair_count": 18, + "serialized_chars": 34769, + "estimated_input_tokens": 8693, + "estimated_eval_tokens": 17642 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Only one true duplicate pair stands out: the two `fixed_cross_entropy` PRs (43251 and 43254). The rest are unrelated model additions, refactors, or tests and should stay separate.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43251 is the clearest representative of the actual fix: it directly addresses the cross-entropy kwargs bug, has an explicit issue link, and makes the narrowest self-contained change in `loss_utils.py`.", + "best_issue_reason": null, + "best_pr_reason": "PR 43251 is the best overall representative because it fixes the concrete bug most explicitly and with the most direct traceability to the reported issue; PR 43254 is effectively the same change but less specific.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "Unrelated features: VideoPrism model/docs vs an ExecuTorch dynamic-shape test. Same broad area of training/runtime infrastructure is not enough." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Mistral tokenizer methods vs ExecuTorch export test; different code paths and different problems." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Tokenizer API cleanup vs adding a new MobileLLM model; unrelated changes." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor-pretrained robustness work vs new MobileLLM model implementation; not the same bug/change." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy kwargs fix vs new MobileLLM model addition; no shared underlying defect." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "Loss utility kwargs fix vs Mistral4 test/model updates; different concrete issues." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "ExecuTorch dynamic-shape test vs Mistral4 test fixes; both are tests but not the same underlying change." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Dynamic-shape export test vs MobileLLM model addition; unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "Model implementation work vs a core class attribute-resolution fix; different code paths and intents." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "New model addition vs Falcon output-collection refactor; not the same change." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor vs a model-creation skill PR; no shared bug or code path." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-interface refactor vs Mistral4 tests/model changes; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 tests/model fixes vs `layrnorm` renaming cleanup; different concrete changes." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 test fixes vs adding a TDT loss kernel; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 test fixes vs config-resolution ordering fix; different code paths." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Model/test fixes vs model-creation skill content; not a duplicate change." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "MobileLLM model implementation vs model-creation skill/docs; unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs modify `src/transformers/loss/loss_utils.py` to fix the same `fixed_cross_entropy` issue by forwarding supported `nn.functional.cross_entropy` kwargs (notably `weight` and `label_smoothing`). They are mergeable into one PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The judgments stay grounded in the packet and the lone accepted duplicate pair is well supported by shared file/issue-target evidence. The other pair rejections are conservative; none look mergeable as a single 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": 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": "926984a8d5e460a648e89ef415d1c57fed30b3445d0140979f17e0d00e71766f", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7226, + "estimated_input_tokens": 1807, + "estimated_eval_tokens": 3870 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7226, + "estimated_input_tokens": 1807, + "estimated_eval_tokens": 3870 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two related VibeVoice PRs, but they target different work: initial VibeVoice support vs. a later realtime/acoustic-tokenizer extension. Not duplicates.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 40546 is the core VibeVoice implementation (model, docs, auto-registry, processing). PR 42781 is a separate realtime follow-up with additional tokenizer/feature-extractor work.", + "best_issue_reason": null, + "best_pr_reason": "PR 40546 is the better cluster anchor because it represents the main VibeVoice integration; 42781 is a distinct extension rather than the same change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "Related but not the same underlying change: 40546 adds base VibeVoice support, while 42781 adds VibeVoice Realtime and acoustic-tokenizer changes. Too distinct to merge as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "a9c041c2bb68f8d6a62eb7ce3c9a258214f1488d3d437c2cc1a47abef4d421b8", + "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 are two unrelated feature pull requests in different model families: TimesFM covariate forecasting vs. DINOv3 image classification support. They should not be merged or triaged as duplicates.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR as a duplicate target: the PRs address different models, code paths, and user-facing capabilities. If one must be chosen as the representative, 41224 is the more narrowly scoped and self-contained change.", + "best_issue_reason": null, + "best_pr_reason": "pull_request:41224 is the best single PR in the cluster because it is focused on one concrete capability addition with a tight implementation surface, while 40755 is a separate larger feature for a different model.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different underlying changes: TimesFM covariate forecasting vs. DINOv3 image classification support. They do not fix the same code-path problem and could not plausibly be merged into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "fa6d7c238f155cb13c29b120743fdb2d0a50caf556bba95fe7a5d8a365bd5947", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10436, + "estimated_input_tokens": 2609, + "estimated_eval_tokens": 5474 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 10436, + "estimated_input_tokens": 2609, + "estimated_eval_tokens": 5474 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are three separate model-integration pull requests (DeepSeek OCR, Isaac, Kimi2-6), not duplicate fixes for the same underlying bug or change. The shared files are just common Transformers boilerplate for new model additions, so the soft links should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true canonical duplicate exists here because the PRs add different models and implement different code paths. If forced to pick one representative, 45630 is the strongest candidate because it has the broadest implementation surface and highest review activity.", + "best_issue_reason": null, + "best_pr_reason": "45630 is the most substantial and reviewed of the three new-model PRs, with extensive auto-mapping and model files plus test coverage, so it is the best representative of this cluster even though it is not semantically the same change as the others.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41797", + "right": "pull_request:45186", + "accept": false, + "reason": "Different model additions: DeepSeek OCR vs Isaac. Shared Transformers scaffolding files do not indicate the same bug or change, and these would not be merged as one PR." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "Different model additions: Isaac vs Kimi2-6. The overlap is generic integration plumbing, not a shared concrete code-path fix or feature." + } + ] + }, + "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": "019d8a8ef99a0fa869b71af98f60cf0bb7d5c71cae52ed2ae02edc2a4b18f90a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5934, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two separate GitHub Actions maintenance PRs, not the same underlying change: one is a broad Node 24 compatibility upgrade for workflow actions, while the other mainly bumps docker/build-push-action and related workflow action versions. The overlap in workflow files is incidental.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No clear canonical duplicate PR: the two PRs address different dependency upgrade goals and should remain separate.", + "best_issue_reason": null, + "best_pr_reason": "PR 42975 is the broader, more central workflow-actions upgrade and is the better representative of the cluster theme, but it is still not a duplicate of 42976.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "Different concrete changes: 42975 targets Node 24 compatibility across GitHub Actions, while 42976 upgrades docker/build-push-action and a smaller set of related workflow actions. Shared workflow paths alone are not enough to treat them as the same PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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-42978-10", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:45487", + "pull_request:45490", + "pull_request:46109", + "pull_request:46135", + "pull_request:46151", + "pull_request:46162", + "pull_request:46214" + ], + "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:45487|pull_request:46214", + "pull_request:45487|pull_request:46151", + "pull_request:45490|pull_request:46162", + "pull_request:46109|pull_request:46214", + "pull_request:46109|pull_request:46151", + "pull_request:46135|pull_request:46214", + "pull_request:46135|pull_request:46151", + "pull_request:46151|pull_request:46214", + "pull_request:46162|pull_request:46214" + ], + "prepared_review_unit_hash": "3a023daa2042dcf1a6d6de47911619e8766eeb251314b097fa39e41c028b1b6d", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 15, + "serialized_chars": 29975, + "estimated_input_tokens": 7494, + "estimated_eval_tokens": 15244 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 15, + "serialized_chars": 29975, + "estimated_input_tokens": 7494, + "estimated_eval_tokens": 15244 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs form two loose themes\u2014new model additions and XPU/test expectation tweaks\u2014but none of the soft edges look like true duplicates of the same concrete change. The similarities are mostly broad-file overlap (docs/auto mappings/tests) rather than shared bug fixes.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR stands out: the set mixes unrelated model additions (Molmo, Molmo2, CTSM, ViT NEPA) with unrelated test-only XPU/distributed updates, so there is no one artifact that safely represents the cluster.", + "best_issue_reason": null, + "best_pr_reason": "No clear global best PR for duplicate-triage purposes; the apparent overlaps are superficial and the PRs address different models or different test paths.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Both are model-addition PRs, but for different model families (ViT NEPA vs Molmo); shared docs/auto files are just repository plumbing." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo2); not the same underlying change." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "Unrelated model additions (ViT NEPA vs CTSM); overlap is only in shared registration/docs files." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are distinct model additions, not one duplicate change." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo vs CTSM are unrelated model adds; shared auto/docs edits are incidental." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM are different models with different codepaths and tests." + }, + { + "left": "pull_request:45487", + "right": "pull_request:46214", + "accept": false, + "reason": "AltCLIP/ChineseCLIP model-parallel fix is unrelated to HRM text XPU expectation updates." + }, + { + "left": "pull_request:45487", + "right": "pull_request:46151", + "accept": false, + "reason": "Different problems: model-parallel device handling vs pipeline token-classification test expectations." + }, + { + "left": "pull_request:45490", + "right": "pull_request:46162", + "accept": false, + "reason": "CTSM model implementation PR vs Florence2 XPU test adjustments; not the same bug or change." + }, + { + "left": "pull_request:46109", + "right": "pull_request:46214", + "accept": false, + "reason": "Both touch tests, but one is distributed trainer parity and the other is HRM text expectations; different concrete test paths." + }, + { + "left": "pull_request:46109", + "right": "pull_request:46151", + "accept": false, + "reason": "Distributed trainer test genericization and token-classification expectations are separate changes." + }, + { + "left": "pull_request:46135", + "right": "pull_request:46214", + "accept": false, + "reason": "Kosmos2.5 XPU expectations and HRM text XPU expectations are model-specific, not a single duplicate fix." + }, + { + "left": "pull_request:46135", + "right": "pull_request:46151", + "accept": false, + "reason": "Different model test suites and different expectation updates; not mergeable as one PR." + }, + { + "left": "pull_request:46151", + "right": "pull_request:46214", + "accept": false, + "reason": "Token-classification pipeline expectations vs HRM text expectations are unrelated model/test updates." + }, + { + "left": "pull_request:46162", + "right": "pull_request:46214", + "accept": false, + "reason": "Florence2 XPU expectations and HRM text XPU expectations are distinct model-specific test changes." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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-43291-4", + "nodes": [ + "pull_request:43291", + "pull_request:43665", + "pull_request:44815", + "pull_request:46102" + ], + "soft_pairs": [ + "pull_request:43291|pull_request:43665", + "pull_request:43291|pull_request:44815", + "pull_request:43665|pull_request:46102", + "pull_request:43665|pull_request:44815" + ], + "prepared_review_unit_hash": "21c978c73e3a32f2fb1de0c1ae4a2239380534c24e5e4aa1a22074200ee57a22", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 9390, + "estimated_input_tokens": 2348, + "estimated_eval_tokens": 4952 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 9390, + "estimated_input_tokens": 2348, + "estimated_eval_tokens": 4952 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are not duplicates: two are test-only edits, one fixes Whisper decoding, one fixes dequant/FP8 loading, and one fixes FSDP tied-weights handling. They touch different code paths and should stay separate.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "If a representative PR is needed, 44815 is the strongest standalone code fix: it changes real loading/quantization logic across multiple files, while the others are either test-only or different subsystems.", + "best_issue_reason": null, + "best_pr_reason": "44815 is the best representative of a substantive bug fix in this cluster; it has concrete implementation changes, unlike 43665 which is test-only and the others which address unrelated areas.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Whisper tokenizer decoding vs CLIP/ViT test edits; different subsystems and no shared bug/change." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper timestamp/batch decode fix is unrelated to FP8 dequant/loading changes." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46102", + "accept": false, + "reason": "Test-only CLIP/ViT assertions are unrelated to FSDP tied-weights logic." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "Both touch tests superficially, but the underlying changes are unrelated: model tests vs quantization/loading code." + } + ] + }, + "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-43488-5", + "nodes": [ + "pull_request:43488", + "pull_request:43532", + "pull_request:45932", + "pull_request:46100", + "pull_request:46114" + ], + "soft_pairs": [ + "pull_request:43488|pull_request:46114", + "pull_request:43488|pull_request:43532", + "pull_request:43532|pull_request:46114", + "pull_request:45932|pull_request:46100", + "pull_request:46100|pull_request:46114" + ], + "prepared_review_unit_hash": "9659f88bb7446b99f073d17e482942180306a7fa696ed903be67ac33d0421692", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 11341, + "estimated_input_tokens": 2836, + "estimated_eval_tokens": 5928 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 11341, + "estimated_input_tokens": 2836, + "estimated_eval_tokens": 5928 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Only 46100 and 46114 are true duplicates: they carry the same one-file ViT test migration diff. The other PRs are test/noise PRs or broader unrelated changes that merely overlap on a filename or subsystem.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46100 is the clean canonical representative: it matches 46114 exactly at the diff level, and its title is the simpler of the duplicate pair.", + "best_issue_reason": null, + "best_pr_reason": "46100 is the best overall PR candidate because it is the clearest exact duplicate target in the cluster and represents the only truly mergeable duplicate pair.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:46114", + "accept": false, + "reason": "Shared ViT test filename is incidental; 43488 changes setup/version/dependencies while 46114 only edits test assertions. Not the same underlying change." + }, + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "These are different fake PRs with different code paths and files; the overlap is too broad and not a concrete duplicate." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46114", + "accept": false, + "reason": "43532 is a workflow/infra change, while 46114 is a local test-file edit. They do not fix the same bug or implement the same change." + }, + { + "left": "pull_request:45932", + "right": "pull_request:46100", + "accept": false, + "reason": "45932 is a broad multi-file model/config/workflow change; 46100 is a narrow ViT test migration. Same broad area, but not the same underlying code change." + }, + { + "left": "pull_request:46100", + "right": "pull_request:46114", + "accept": true, + "reason": "Exact same diff in tests/models/vit/test_modeling_vit.py; these are the same test migration change and could plausibly be merged as one PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet. The only accepted soft edge, 46100/46114, is supported by identical diff previews in the same file, and the rejected pairs are conservative given the broader or unrelated changes. Canonical PR 46100 is a reasonable duplicate representative." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "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": "5a07b8df0a7e47e3c115639cd3ee798ee1184dd2927552ec18ce7b21f4180cc8", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6083, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6083, + "estimated_input_tokens": 1521, + "estimated_eval_tokens": 3298 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are closely related in theme and documentation scaffolding, but they implement different pipelines for different segmentation tasks and model integrations. They should not be treated as duplicates.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: 43612 adds Promptable Concept Segmentation (SAM3-focused), while 43613 adds Promptable Visual Segmentation with different model hooks and test coverage.", + "best_issue_reason": null, + "best_pr_reason": "43613 is the better representative PR for the cluster because it is the broader, more fully integrated pipeline addition; 43612 is a separate narrower feature.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Reject: they share pipeline/docs boilerplate, but the underlying changes are different features with different code paths and model/test integrations (concept segmentation vs visual segmentation)." + } + ] + }, + "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": "17aca8c302f0419d5d93e8cfcb5c87d1718c26f9a34f6322cf81a2d93b29e603", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4292, + "estimated_input_tokens": 1073, + "estimated_eval_tokens": 2402 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated trainer.py pull requests: one adds `_loss_is_scaled_for_ga` for gradient-accumulation loss scaling, the other extracts `_get_signature_columns` for column filtering. Same file, different code paths, not the same underlying change.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the patches solve different Trainer extension-point problems and are not mergeable into one coherent fix.", + "best_issue_reason": null, + "best_pr_reason": "Neither PR is a good global representative for the cluster because they address unrelated behaviors in `Trainer` despite touching the same file.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Different functionality and different code paths: one changes gradient-accumulation loss normalization, the other refactors signature-column inspection for data filtering." + } + ] + }, + "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-43656-2", + "nodes": [ + "pull_request:43656", + "pull_request:43842" + ], + "soft_pairs": [ + "pull_request:43656|pull_request:43842" + ], + "prepared_review_unit_hash": "f60eca232ef0b4e630eb6a8a5365ab10bae824d47494678849c386f5b5098c28", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3664, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 3664, + "estimated_input_tokens": 916, + "estimated_eval_tokens": 2088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs touch the same file and both address annotation-related NameErrors in `src/transformers/cli/serve.py`, but they fix different runtime paths: one is a broad typing/future-annotations cleanup, the other is specifically about `TypeAdapter` when `pydantic` is absent.", + "confidence": 0.87, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #43656 is the broader representative fix for import-time typing/annotation errors in the CLI, with more review/discussion and a wider cleanup than the narrower `TypeAdapter`-specific patch.", + "best_issue_reason": null, + "best_pr_reason": "PR #43656 most centrally captures the cluster\u2019s underlying serve.py annotation problem and is the more comprehensive change set.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Related but not duplicates: #43656 adds `from __future__ import annotations` and broad type-hint cleanup, while #43842 specifically guards/quotes `TypeAdapter` for the no-pydantic case. They address different concrete NameError paths and could be merged separately." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 13, + "component_count": 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:44018|pull_request:44066", + "pull_request:44066|pull_request:44068", + "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:44018|pull_request:44116", + "pull_request:44068|pull_request:44722", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44086", + "pull_request:44018|pull_request:44085", + "pull_request:44068|pull_request:44086", + "pull_request:44071|pull_request:44085", + "pull_request:44059|pull_request:44086", + "pull_request:44085|pull_request:44086", + "pull_request:44098|pull_request:44101", + "pull_request:44059|pull_request:44722", + "pull_request:44018|pull_request:44071", + "pull_request:44017|pull_request:44722", + "pull_request:44017|pull_request:44161", + "pull_request:44017|pull_request:44073", + "pull_request:44068|pull_request:44071", + "pull_request:44071|pull_request:44086", + "pull_request:43997|pull_request:44073", + "pull_request:44017|pull_request:44071", + "pull_request:44073|pull_request:44101", + "pull_request:44059|pull_request:44071", + "pull_request:44018|pull_request:44059", + "pull_request:44073|pull_request:44161", + "pull_request:44056|pull_request:44101", + "pull_request:44059|pull_request:44068", + "pull_request:44101|pull_request:44161", + "pull_request:44017|pull_request:44018", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request:43997|pull_request:44018", + "pull_request:44056|pull_request:44161", + "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:44010|pull_request:44018", + "pull_request:44071|pull_request:44129", + "pull_request:43997|pull_request:44129", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154" + ], + "prepared_review_unit_hash": "274d4d38f118395514d4dbd5706eeed927fc123e9abf8e28d63166c45d66481d", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84899, + "estimated_input_tokens": 21225, + "estimated_eval_tokens": 42706 + }, + "budget": { + "node_count": 27, + "item_count": 27, + "soft_pair_count": 72, + "serialized_chars": 75096, + "estimated_input_tokens": 18774, + "estimated_eval_tokens": 37804 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a wave of standardized output-tracing refactors across many model backbones. Many of the relevant PRs point to #43979, but several do not, so it\u2019s better framed as a broad theme plus a GPT-style subset where same-file duplicates are plausible.", + "confidence": 0.87, + "canonical_issue_reason": "#43979 is the shared umbrella tracker for the output-tracing modernization work in the relevant PRs, though not every PR in the cluster explicitly links it.", + "canonical_pr_reason": "PR 44722 is the clearest representative because it explicitly targets #43979 and captures the core GPT-J/CodeGen output-tracing refactor in the same code path as the GPT-style duplicates.", + "best_issue_reason": "#43979 is the best cluster anchor because it ties together the main standardized output-tracing refactor wave, especially the GPT-style subset.", + "best_pr_reason": "PR 44722 is the best single representative: it is explicit about the shared tracker, spans the core GPT-J/CodeGen implementation, and best summarizes the duplicate-prone subset.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different model families and code paths; shared output-tracing theme is not enough." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "Different backbones and refactor scopes; same tracker/theme alone does not make them duplicates." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing code path in the same file; this looks like an overlapping slice of one refactor." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "ResNet/RT-DETR ResNet versus EfficientNet are different model implementations, so these are separate changes." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet and GPT-J/CodeGen are unrelated code paths; only the refactor pattern matches." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "GPT-J/CodeGen versus EfficientNet are different implementations and should not be merged as one PR." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTa are unrelated model-specific refactors." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen and MGP-STR are different model families and code paths." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen versus MPT are distinct implementations; the shared decorator pattern is not enough." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Same GPT-Neo output-tracing refactor in the same model/file; these plausibly duplicate the same change." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model implementations." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen and GPT-Neo are different code paths, so these are not duplicates." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet and GPT-J/CodeGen are unrelated backbones despite similar refactor wording." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are separate model-specific changes." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet and GPT-J are unrelated implementations; same theme is not enough." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are different model families and not the same concrete change." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet and MPT are separate model-specific refactors." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert/XLM are different code paths; only the refactor style is shared." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo versus GPT-J/CodeGen are different implementations, so this is not a duplicate." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44722", + "accept": false, + "reason": "MPT and GPT-J/CodeGen are unrelated model families." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44059", + "accept": false, + "reason": "GPT-Neo and GPT-2 are separate model implementations and should stay separate." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are unrelated apart from the shared output-tracing pattern." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert/XLM are different model families with distinct code paths." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44722", + "accept": false, + "reason": "SegFormer and GPT-J/CodeGen are unrelated model implementations." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44161", + "accept": false, + "reason": "SegFormer and LongT5 are different architectures; only the refactor style matches." + }, + { + "left": "pull_request:44017", + "right": "pull_request:44073", + "accept": false, + "reason": "SegFormer and VisualBERT are different code paths and not duplicates." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different model-specific output-tracing changes." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-2 and MGP-STR are separate model implementations." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J file and same decorator-based refactor; 44722 just broadens coverage with the same underlying GPT-J change." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are unrelated code paths; the shared tracker does not make them duplicates." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "Different model families and different forward/output plumbing changes." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are unrelated implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are different model families with separate refactors." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and VITS are unrelated model-specific output-tracing changes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen output-tracing refactor in the same code path; 44722 looks like the broader overlapping iteration." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J output-tracing change in the same file; 44722 incorporates that change and extends it to CodeGen." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are unrelated code paths." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "Different model families and different output-handling implementations." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are unrelated model implementations." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTa and SpeechT5 are distinct model families." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTa and VITS are different code paths; same refactor pattern only." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The cluster theme is mostly grounded, but the summary overstates the evidence by saying all PRs are \"orbiting\" #43979; several items in the packet have no explicit issue link, so that should be softened to \"many/most\" or \"the relevant PRs.\" The soft-edge verdicts are generally conservative, and the same-model duplicate calls look plausible, but I would still tighten the umbrella claim before accepting." + }, + "evaluator_used": true, + "retried": true, + "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": "f38a7ba969b9d7759b3af9963b1a1d2f9ab5d677b8533c1d9bdac70f3c509405", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 78, + "serialized_chars": 84899, + "estimated_input_tokens": 21225, + "estimated_eval_tokens": 42706 + }, + "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 related to a common standardized output-tracing refactor pattern, but they touch different model implementations and distinct code paths, so they are not duplicates of one another. I would treat them as separate PRs in the same thematic cluster.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "44129 is the strongest representative because it has the clearest scoped refactor, an explicit linked issue target, and concrete output-tracing changes in one model. That said, it is still not a duplicate of the others.", + "best_issue_reason": null, + "best_pr_reason": "44129 is a good cluster representative since it is a focused model-specific implementation of the shared output-capture refactor theme and includes an explicit issue reference.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "Both are standardized-output refactors, but one is for FocalNet and the other for ImageGPT; they change different model code paths and are not mergeable as 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. The similarity is the refactor pattern, not the underlying bug/change in a single code path." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "These target different models (MPNet vs VisualBERT) and different tracing hooks, so they are not the same concrete fix." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "MPNet and TextNet are unrelated model files with separate output-tracing refactors; same theme, different changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBERT are distinct model-specific migrations to the shared output-capture pattern, not the same underlying bug or change." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "SqueezeBERT and SpeechT5 modify different model forward paths. Sharing the standardized output-capture goal is not enough to treat them as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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-43998-14", + "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", + "pull_request:46106" + ], + "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", + "pull_request:44027|pull_request:46106", + "pull_request:44026|pull_request:46106", + "pull_request:44001|pull_request:46106", + "pull_request:44002|pull_request:46106", + "pull_request:43998|pull_request:46106" + ], + "prepared_review_unit_hash": "bfbd37308589ee199874faa3eaa56ea422b593f808c719179147d4323e46ccd4", + "component_budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 45, + "serialized_chars": 43783, + "estimated_input_tokens": 10946, + "estimated_eval_tokens": 22148 + }, + "budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 45, + "serialized_chars": 43783, + "estimated_input_tokens": 10946, + "estimated_eval_tokens": 22148 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is a set of separate model-specific PRs that all follow the same output-tracing refactor pattern, but they do not fix the same concrete code path. I reject the soft duplicate edges and treat them as distinct PRs. PR 44025 is the best representative of the cluster.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 44025 is the clearest representative: it is a self-contained output-tracing refactor, spans two closely related files, and directly reflects the cluster\u2019s common migration pattern.", + "best_issue_reason": null, + "best_pr_reason": "PR 44025 is the best single PR to anchor the cluster because it is explicit, focused, and representative without being overly specialized to one unusual code path.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model implementations and forward paths; same refactor theme, but not the same mergeable change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different architectures (`timm_backbone` vs `vision_text_dual_encoder`); similar output-tracing cleanup only." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Separate model code paths; not the same underlying bug or fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Both are output-tracing refactors, but they target different model families and concrete implementations." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Unrelated model-specific forward paths; too broad to be one duplicate PR." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "Different model code paths (`mobilenet_v1` vs `univnet`); only the refactor style overlaps." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model families and outputs; not the same concrete change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "Separate models with distinct forward logic; same tracing migration, not one PR-worthy fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Different model code paths; the similarity is only the output-tracing refactor pattern." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "No shared underlying bug or code path beyond the general refactor theme." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Different architectures (`vision_text_dual_encoder` vs `univnet`); not mergeable as one fix." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model implementations; only a shared migration approach." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "Distinct model families and forward methods; not the same concrete problem." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Both touch output handling, but on unrelated model code paths." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "Separate models and outputs; same refactor pattern is not enough to duplicate them." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model stacks (`univnet` vs `falcon_mamba/mamba`); not one underlying fix." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "Different code paths and different model families; only broadly similar tracing cleanup." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model implementations; not the same bug or change." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Separate model code paths; same general refactor class only." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model families (`mamba` vs `superpoint`) and distinct output structures." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Codegen refactor is a different concrete change than the Mamba/Falcon-Mamba tracing work." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model families and forward paths; similar refactor intent only." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "Different code paths (`univnet` vs `codegen`); not a duplicate change." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model-specific output handling; no single shared bug to merge." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "Different models (`depth_anything` vs `vision_encoder_decoder`) and different forward logic." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "Different model wrappers and output signatures; same refactor pattern only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "Unrelated model code paths; not the same underlying change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Different architectures and output plumbing; too broad to count as one duplicate." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model code paths; only the output-tracing migration overlaps." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and forward implementations; not one fix." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "Different encoder-decoder wrappers; same broad refactor theme, but not the same code path." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model implementations and output types; not mergeable into one PR." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "Distinct model code paths; shared tracing refactor only." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "Speech encoder-decoder and SuperPoint are unrelated concrete changes." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and output handling; not a duplicate." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "Separate model-specific forward paths; only the refactor style is similar." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "Different models and outputs; no shared underlying bug." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "Unrelated model code paths; same broad output-tracing refactor only." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "Different model families and internal logic; not the same concrete fix." + }, + { + "left": "pull_request:44027", + "right": "pull_request:46106", + "accept": false, + "reason": "Documentation/toctree change is unrelated to the speech encoder-decoder tracing refactor." + }, + { + "left": "pull_request:44026", + "right": "pull_request:46106", + "accept": false, + "reason": "Docs toctree cleanup is not the same as the vision-encoder-decoder code change." + }, + { + "left": "pull_request:44001", + "right": "pull_request:46106", + "accept": false, + "reason": "Unrelated docs navigation edit versus a model tracing refactor." + }, + { + "left": "pull_request:44002", + "right": "pull_request:46106", + "accept": false, + "reason": "Documentation tree update is unrelated to UperNet output tracing." + }, + { + "left": "pull_request:43998", + "right": "pull_request:46106", + "accept": false, + "reason": "Docs-only toctree change is unrelated to the timm_backbone refactor." + } + ] + }, + "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": "97016c49c9930f8896c30fdcf6ff36a2f98b26e014812c5b3364d146e0126788", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10912, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "All three PRs are standalone model-addition changes for different models (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They share only framework boilerplate and auto-registration patterns, not the same underlying feature or code-path fix.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "No true canonical PR: these are separate model implementation PRs for different architectures, so there is no duplicate to collapse into one.", + "best_issue_reason": null, + "best_pr_reason": "PR 45075 is the strongest standalone representative: it has the broadest implementation surface (core model, processing, auto mappings, tests, docs, and conversion helpers) and a clear, self-contained scope.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different models and different functionality (PenguinVL vs Voxtral TTS). Shared files are only registration/docs boilerplate, not the same bug or feature." + }, + { + "left": "pull_request:44662", + "right": "pull_request:45075", + "accept": false, + "reason": "Both add model support, but for unrelated models with different code paths and artifacts; this is broad subsystem overlap, not a duplicate change." + }, + { + "left": "pull_request:45075", + "right": "pull_request:45401", + "accept": false, + "reason": "DeepSeek-OCR-2 and Voxtral TTS are distinct model integrations with different processing and modeling logic; they are not plausibly mergeable into one PR as the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "e6d9a11e0ed8bd4021ac41e680dc1615b7bf9fedf589a013bdd75ea25af1680c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4408, + "estimated_input_tokens": 1102, + "estimated_eval_tokens": 2460 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are related in the broad area of module/kernel fusion, but they are not the same change: one adds a standalone Module Fusion API scaffold, while the other wires up n-to-1 kernel fusion through KernelConfig and integration points. I would not treat them as duplicates.", + "confidence": 0.86, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45363 is the better canonical PR because it is the more complete, user-facing feature implementation with multiple integration changes, whereas #44979 is mainly the lower-level Module Fusion API foundation.", + "best_issue_reason": null, + "best_pr_reason": "PR #45363 is the best representative of the cluster\u2019s overall theme since it delivers the broader kernel-fusion workflow; #44979 is supporting infrastructure rather than the same concrete change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Related fusion infrastructure, but not the same concrete bug/change and not something that should be merged into a single PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "724597b27a3cef28d0c2128dc1553b048b6911a327270a16447c8982e6183007", + "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 address the same Mistral regex patching AttributeError in the same code path and target the same issue; 45317 is the more complete variant because it adds tests.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #45081 is the shared explicit target for both PRs, so it is the natural canonical issue.", + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same fix plus test coverage, making it the more complete and likely final form of the change.", + "best_issue_reason": "#45081 is the only explicit issue target and represents the underlying bug both PRs are trying to resolve.", + "best_pr_reason": "pull_request:45317 is the best representative PR: same fix, same file, and it adds tests to validate the AttributeError regression.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying AttributeError fix in _patch_mistral_regex, same target issue #45081, and the diffs overlap on the exact code-path; 45317 is a stricter superset with tests, so these are duplicate/same-change PRs." + } + ] + }, + "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": "1b71047d5bf36b9aca79b73d2fb8e0308e74f8b85cecaad6793215abd9f32f6d", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6066, + "estimated_input_tokens": 1517, + "estimated_eval_tokens": 3290 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 6066, + "estimated_input_tokens": 1517, + "estimated_eval_tokens": 3290 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are related to NemotronH internals, but they address different changes: one is a refactor/inheritance and checkpoint conversion mapping update, while the other introduces the dense/sparse model split with docs, auto-mapping, and tests.", + "confidence": 0.89, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45569 is the better canonical representative because it is the broader, more complete feature change: it adds the new NemotronH dense/sparse model structure, updates docs and auto-mappings, and includes tests. PR 45115 is a narrower refactor focused on GraniteMoeHybrid-style inheritance and conversion renames.", + "best_issue_reason": null, + "best_pr_reason": "PR 45569 is the strongest standalone PR in the cluster since it carries the main user-visible/model-structure work and supporting documentation/tests, whereas PR 45115 is a narrower implementation refactor.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "Reject: despite overlapping NemotronH files, the PRs are not the same underlying change. 45115 is about refactoring NemotronH to inherit GraniteMoeHybrid and adding conversion-key renames, while 45569 is about introducing NemotronH dense/sparse variants plus docs, auto-mapping, and tests. They could coexist, but they are not one mergeable duplicate PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "2c1454679a95544862b9541330e98e2cec70c98157195ee5865798e5ca03dcc6", + "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 but distinct feature PRs: one adds heterogeneous config support, the other adds heterogeneous model/runtime support. They overlap in shared config helpers, but they do not look like the same mergeable change.", + "confidence": 0.79, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45332 is the more canonical umbrella change: it implements the concrete model/runtime behavior for heterogeneous layers and includes the cache/modeling updates that depend on config support.", + "best_issue_reason": null, + "best_pr_reason": "PR #45332 is the strongest standalone candidate because it covers the main end-user behavior change, not just the supporting config plumbing.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "These are related layers of the same feature area, but one is config plumbing and the other is model/runtime support; they do not look like the same concrete code-path problem that could be merged into a single PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "88128e6ef8b67744d07c06e7f912a37f00472f2d4c421b82781fa87f5f88c598", + "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-focused cleanups, but they are not the same concrete change: #45415 is a broad repo-wide type-checking pass with pyproject typing-ignore adjustments, while #45425 adds specific typing helper protocols/aliases and related modeling-utils annotations. They share some files, but the underlying work is different enough that they should not be merged as duplicates.", + "confidence": 0.79, + "canonical_issue_reason": null, + "canonical_pr_reason": "#45415 is the broader umbrella typing cleanup and best represents the cluster\u2019s overall theme; #45425 is a narrower, more specialized follow-up-style typing change.", + "best_issue_reason": null, + "best_pr_reason": "#45415 is the best representative PR because it covers the widest scope of the typing initiative and includes the typing-policy change in pyproject.toml, making it the most canonical cluster member.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Shared typing-related files are not enough: #45415 is a large multi-file type-checking sweep, while #45425 introduces new typing protocols/aliases for specific internals. They do not fix the same concrete bug or merge cleanly into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45513-3", + "nodes": [ + "pull_request:45513", + "pull_request:46084", + "pull_request:46098" + ], + "soft_pairs": [ + "pull_request:45513|pull_request:46084", + "pull_request:46084|pull_request:46098" + ], + "prepared_review_unit_hash": "ce1bf896e68f5dd02102fd168710da03ac3ce91c513e0de8852e7c74b5c8f9d1", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 7427, + "estimated_input_tokens": 1857, + "estimated_eval_tokens": 3970 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 7427, + "estimated_input_tokens": 1857, + "estimated_eval_tokens": 3970 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two Mamba2 PRs appear to address the same use_cache seq_len>1 bug; the Qwen3.5 PR is unrelated and should stay separate. The broader Mamba2 PR is the best representative of that fix.", + "confidence": 0.78, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46098 is the best canonical representative because it directly targets the Mamba2 use_cache seq_len>1 incorrect-results bug, covers both CPU/GPU paths, and includes tests/docs, making it the most complete version of the fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #46098 is the strongest cluster representative: it is explicit about the underlying bug, has the same Mamba2 code path, and is broader/more complete than the narrower GPU-focused predecessor PR #46084.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45513", + "right": "pull_request:46084", + "accept": false, + "reason": "Different model families and different implementations; they only share a generic cached-forward symptom, not the same concrete bug or mergeable change." + }, + { + "left": "pull_request:46084", + "right": "pull_request:46098", + "accept": true, + "reason": "Both target the same Mamba2 use_cache seq_len>1 bug in modeling_mamba2.py; #46098 looks like a broader follow-up/superset with the same underlying fix and could plausibly be merged into one PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: PR #45513 is a different model family and does not look mergeable with the Mamba2 fix, while PRs #46084 and #46098 share the same issue target and Mamba2 files, so treating #46098 as the broader canonical representative is conservative. No evidence is overstated." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "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": "738cdf6fadb059a2922d512c4a817a4753d3e7ad472b9fad9596fb34b13a705e", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs touch expert-parallel loading, but they fix different bugs: one addresses missing distributed config/EP detection for ZeRO-3 + accelerate launch, while the other fixes experts being overwritten during FSDP2 sharding/broadcast. They are related but not duplicates.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45662 is the better canonical representative: it targets a concrete, specific code-path failure (expert weights being overwritten during distributed loading) and includes the broader set of loading-path changes needed for that bug.", + "best_issue_reason": null, + "best_pr_reason": "PR #45662 is the strongest standalone PR in the cluster because it fixes a clearly defined runtime bug with broader implementation coverage; #45548 is narrower and more about enabling EP metadata/detection for a different loading path.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Different underlying bugs and fix paths: #45548 is about EP + DeepSpeed ZeRO-3 loading via accelerate launch and serialization of distributed_config/has_ep, while #45662 fixes EP + FSDP2 experts being silently overwritten by rank-0 broadcast via post-shard wrapping. Shared subsystem only, not the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 24, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45639-2", + "nodes": [ + "pull_request:45639", + "pull_request:45645" + ], + "soft_pairs": [ + "pull_request:45639|pull_request:45645" + ], + "prepared_review_unit_hash": "b4d3d2474892b2c331ec845fb57ffdca67449743d433224688b3ddf66e45b557", + "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 collision around `captured_info`/patched testing debug output, but PR 45645 is the broader superset that also updates CI log collection. PR 45639 looks like the narrower core fix; 45645 is the better canonical PR.", + "confidence": 0.92, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45645 is the most complete version of the fix: it includes the xdist-safe output handling in `testing_utils.py` plus the CI/workflow changes needed to collect multiple `captured_info*.txt` files, so it subsumes PR 45639.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the best representative because it covers the same underlying xdist collision bug while also preserving downstream CI/debug-log behavior, making it the more comprehensive and merge-ready change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Accept: both PRs fix the same xdist collision around `captured_info` artifacts in `testing_utils.py`; 45645 extends the same fix with workflow/log-handling updates, so they are close enough to be merged as one change set." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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-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": 26, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46069-4", + "nodes": [ + "pull_request:46069", + "pull_request:46073", + "pull_request:46154", + "pull_request:46196" + ], + "soft_pairs": [ + "pull_request:46069|pull_request:46073", + "pull_request:46069|pull_request:46154", + "pull_request:46069|pull_request:46196", + "pull_request:46073|pull_request:46154", + "pull_request:46073|pull_request:46196", + "pull_request:46154|pull_request:46196" + ], + "prepared_review_unit_hash": "c4bd2cb23e1dad30f2521a0881b22dbffcb724be34b6d1680dccf93e6cd46d50", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 10497, + "estimated_input_tokens": 2625, + "estimated_eval_tokens": 5506 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 10497, + "estimated_input_tokens": 2625, + "estimated_eval_tokens": 5506 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are all about adding SDPA / attn_implementation dispatch support, but each one targets a different model family and code path. They are similar feature PRs, not true duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: each PR implements the same feature pattern in a different model-specific attention path, so they should remain separate.", + "best_issue_reason": null, + "best_pr_reason": "PR 46196 is the strongest representative of the cluster because it covers two closely related model variants (SeamlessM4T and SeamlessM4T-v2) and has the broadest attention-dispatch changes.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46069", + "right": "pull_request:46073", + "accept": false, + "reason": "Same broad SDPA support theme, but ShieldGemma2 and CTRL are different model implementations with different attention code paths." + }, + { + "left": "pull_request:46069", + "right": "pull_request:46154", + "accept": false, + "reason": "Both add attention-dispatch support, but ShieldGemma2 and T5 are unrelated architectures and the fixes are model-specific." + }, + { + "left": "pull_request:46069", + "right": "pull_request:46196", + "accept": false, + "reason": "Different model families and files; this is the same feature pattern, not the same concrete bug/change." + }, + { + "left": "pull_request:46073", + "right": "pull_request:46154", + "accept": false, + "reason": "CTRL and T5 each need separate dispatch plumbing; they are not the same underlying code-path fix." + }, + { + "left": "pull_request:46073", + "right": "pull_request:46196", + "accept": false, + "reason": "These are separate model-specific SDPA integrations, not a single change that could plausibly be merged into one PR." + }, + { + "left": "pull_request:46154", + "right": "pull_request:46196", + "accept": false, + "reason": "T5 and SeamlessM4T/v2 are different implementations; similarity is limited to the shared attention-dispatch feature." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 27, + "component_count": 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-46152-2", + "nodes": [ + "pull_request:46152", + "pull_request:46207" + ], + "soft_pairs": [ + "pull_request:46152|pull_request:46207" + ], + "prepared_review_unit_hash": "47ebff8889b0589ea55af2a3ba2af351eb6425439cc43afba987a7fe4e94f938", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4269, + "estimated_input_tokens": 1068, + "estimated_eval_tokens": 2392 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4269, + "estimated_input_tokens": 1068, + "estimated_eval_tokens": 2392 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs aim to remove file-open/source-reading hacks, but they fix different code paths in different modules and are not mergeable as one change.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46152 is the broader umbrella fix: it updates import-time path resolution in `import_utils.py` to avoid reading the module\u2019s own source file, which matches the cluster\u2019s shared theme most directly.", + "best_issue_reason": null, + "best_pr_reason": "PR #46152 is the better cluster representative because its title and diff capture the general pattern of eliminating self-source-file reads, while #46207 is a separate, narrower introspection refactor in `modeling_utils.py`.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46152", + "right": "pull_request:46207", + "accept": false, + "reason": "Related pattern, but not the same underlying bug/change: one rewrites resource-path resolution in `import_utils.py`, the other changes attention-implementation capability detection in `modeling_utils.py`. They touch different subsystems and would not plausibly be one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/manifest.json b/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..97253b561d5851af83f268b3af19802b405ebbf8 --- /dev/null +++ b/snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260526t120027z", + "artifacts": { + "hybrid": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-05-26T12:14:05Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260526T120027Z", + "variant": "hybrid" +} diff --git a/snapshots/20260526T120027Z/manifest.json b/snapshots/20260526T120027Z/manifest.json index c74c4528996f26bcc00ae32e3474f9c398a9bd8f..e8e82bc58ef50c05b365d112e289ca155a1dffa0 100644 --- a/snapshots/20260526T120027Z/manifest.json +++ b/snapshots/20260526T120027Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 1621 }, "extracted_at": "2026-05-26T12:00:27Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260526t120027z", + "runs": { + "hybrid-model-20260526t120027z": { + "analysis_id": "hybrid-model-20260526t120027z", + "artifacts": { + "hybrid": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260526T120027Z/analysis-runs/hybrid-model-20260526t120027z/manifest.json", + "model": null, + "published_at": "2026-05-26T12:14:05Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260526T120027Z", "watermark": {