diff --git a/analysis/current/analysis-report-hybrid.json b/analysis/current/analysis-report-hybrid.json index 88a657880201bfb356271b10afcc35aa4be56306..f5d79c384b7a6be49454400baf53c28196480091 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": "20260522T000112Z", - "generated_at": "2026-05-22T00:05:46Z", + "snapshot_id": "20260522T060021Z", + "generated_at": "2026-05-22T06:10:04Z", "evidence_quality": "full", "llm_enrichment": true, "meta_bugs": [ @@ -1708,133 +1708,6 @@ } ] }, - { - "cluster_id": "cluster-44001-3", - "summary": "Cluster of 3 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 44002, - "issue_numbers": [], - "pr_numbers": [ - 44001, - 44002, - 44004 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 44001, - "right_pr_number": 44002, - "code_similarity": 0.175, - "size_similarity": 0.595, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.375, - "shared_filenames": [], - "shared_file_areas": [] - }, - { - "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": [] - }, - { - "left_pr_number": 44002, - "right_pr_number": 44004, - "code_similarity": 0.125, - "size_similarity": 0.398, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.304, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, - { - "cluster_id": "cluster-43656-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 43656, - "issue_numbers": [], - "pr_numbers": [ - 43656, - 43842 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 43656, - "right_pr_number": 43842, - "code_similarity": 0.405, - "size_similarity": 0.5, - "file_overlap": 1.0, - "area_overlap": 0.0, - "patch_similarity": 0.036, - "shared_filenames": [ - "src/transformers/cli/serve.py" - ], - "shared_file_areas": [ - { - "filename": "src/transformers/cli/serve.py", - "left_ranges": [ - [ - 11, - 18 - ], - [ - 30, - 36 - ], - [ - 315, - 323 - ], - [ - 665, - 671 - ], - [ - 931, - 937 - ], - [ - 1843, - 1849 - ], - [ - 1868, - 1874 - ] - ], - "right_ranges": [ - [ - 54, - 61 - ], - [ - 587, - 593 - ] - ] - } - ] - } - ] - }, { "cluster_id": "cluster-46100-2", "summary": "Cluster of 2 related pull requests linked by soft_similarity.", @@ -1882,35 +1755,6 @@ } ] }, - { - "cluster_id": "cluster-44000-2", - "summary": "Cluster of 2 related pull requests linked by soft_similarity.", - "status": "open", - "confidence": 0.5, - "canonical_issue_number": null, - "canonical_pr_number": 44025, - "issue_numbers": [], - "pr_numbers": [ - 44000, - 44025 - ], - "evidence_types": [ - "soft_similarity" - ], - "pr_comparisons": [ - { - "left_pr_number": 44000, - "right_pr_number": 44025, - "code_similarity": 0.206, - "size_similarity": 0.78, - "file_overlap": 0.0, - "area_overlap": 0.0, - "patch_similarity": 0.333, - "shared_filenames": [], - "shared_file_areas": [] - } - ] - }, { "cluster_id": "cluster-44018-2", "summary": "Cluster of 2 related pull requests linked by soft_similarity.", @@ -2216,15 +2060,6 @@ "target_issue_number": 43240, "reason": "PRs in cluster-43240-3 are treated as duplicates because they converge on issue #43240 with closing_reference, shared_issue_target evidence." }, - { - "cluster_id": "cluster-43656-2", - "canonical_pr_number": 43656, - "duplicate_pr_numbers": [ - 43842 - ], - "target_issue_number": null, - "reason": "PRs in cluster-43656-2 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-43698-3", "canonical_pr_number": 43779, @@ -2251,25 +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-44000-2", - "canonical_pr_number": 44025, - "duplicate_pr_numbers": [ - 44000 - ], - "target_issue_number": null, - "reason": "PRs in cluster-44000-2 are treated as duplicates because they share soft_similarity evidence." - }, - { - "cluster_id": "cluster-44001-3", - "canonical_pr_number": 44002, - "duplicate_pr_numbers": [ - 44001, - 44004 - ], - "target_issue_number": null, - "reason": "PRs in cluster-44001-3 are treated as duplicates because they share soft_similarity evidence." - }, { "cluster_id": "cluster-44018-2", "canonical_pr_number": 44068, diff --git a/analysis/current/analysis-report-hybrid.llm-reviews.json b/analysis/current/analysis-report-hybrid.llm-reviews.json index b7dd4e209e71fba9ba543b1f2be3e0c2b765afe5..069af027274340ffbb3ab609b4006a8e68da95b8 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": "20260522T000112Z", - "generated_at": "2026-05-22T00:05:46Z", + "snapshot_id": "20260522T060021Z", + "generated_at": "2026-05-22T06:10:04Z", "model": "gpt-5.4-mini?service_tier=flex", "reviews": [ { @@ -13,8 +13,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -135,352 +135,412 @@ "issue:39290|issue:43531", "issue:41093|issue:43638" ], - "prepared_review_unit_hash": "1d630c83475b4336c3d89ca71a37cd3a48c2fbde2a21da60f265fae9d5ea87f7", + "prepared_review_unit_hash": "292d8f46e520f108716c719dfd9801e6aee773095874bda6d2225fe08bae8274", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 66, - "serialized_chars": 47529, - "estimated_input_tokens": 11883, - "estimated_eval_tokens": 24022 + "serialized_chars": 47536, + "estimated_input_tokens": 11884, + "estimated_eval_tokens": 24024 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly unrelated issue reports, with one clear duplicate cluster around DeepSpeed ZeRO-3 index-0 out-of-bounds failures and one likely AutoTokenizer v5 regression cluster. The rest are distinct model- or docs-specific problems.", - "confidence": 0.71, - "canonical_issue_reason": "Issue 43638 is the broadest representative of the clearest duplicate cluster: it describes the same DeepSpeed ZeRO-3 index-0 out-of-bounds failure without being tied to a single model, so it best subsumes the more specific reports.", + "summary": "This cluster is mostly heterogeneous and the soft-similarity links are largely false positives. The only clear duplicate pockets are the DeepSpeed ZeRO-3 Bert index error pair and the AutoTokenizer v5 token-ID regression pair.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 45920 is the broadest, most representative report of the shared AutoTokenizer v5 token-ID regression, and it subsumes the narrower Granite-only report in 45812.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43638 is the best canonical issue here because it captures the shared ZeRO-3 failure in the most general terms and can absorb the closely related reports; it is clearer and less model-specific than the others.", + "best_issue_reason": "45920 best captures the underlying bug because it describes the regression across multiple models, making it the most useful umbrella issue.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44322", "right": "issue:45464", "accept": false, - "reason": "Different failures: config attribute missing versus streaming chat/completions error for a model family; not the same code path." + "reason": "Different failures: Qwen3.5 config attribute error vs streaming chat/completions failure." }, { "left": "issue:45237", "right": "issue:45685", "accept": false, - "reason": "Unrelated platforms and bugs: AMD GPU support for GPT-OSS versus an mps histogram op issue in moe code." + "reason": "Unrelated model/runtime issues; AMD GPU support is not the same bug as the MPS histogram error." }, { "left": "issue:43716", "right": "issue:45237", "accept": false, - "reason": "Image preprocessor dtype mismatch is unrelated to GPT-OSS failing on AMD GPUs." + "reason": "Both are device/dtype-related, but they affect different models and code paths." }, { "left": "issue:43828", "right": "issue:45237", "accept": false, - "reason": "Different model and failure mode; autocast dtype mismatch is not the same as the AMD GPU issue." + "reason": "Autocast dtype mismatch and AMD GPU incompatibility are not the same underlying problem." }, { "left": "issue:45237", "right": "issue:45507", "accept": false, - "reason": "Different bug classes: AMD GPU runtime failure versus an invalid method call in GraniteMoEHybrid." + "reason": "Different model families and different failure modes." }, { "left": "issue:44292", "right": "issue:45237", "accept": false, - "reason": "Distinct model/runtime issues; Qwen-3 NVFP4 running error is not the same as GPT-OSS AMD GPU failure." + "reason": "Qwen-3 NVFP4 failure is unrelated to the AMD GPU issue." }, { "left": "issue:30277", "right": "issue:35545", "accept": false, - "reason": "Different subsystems and symptoms: DeepSpeed collective mismatch versus ONNX export failure." + "reason": "DeepSpeed collective mismatch and ONNX export failure are separate bugs." }, { "left": "issue:35545", "right": "issue:42915", "accept": false, - "reason": "Both are model-related, but one is ONNX export and the other is a FineGrainedFP8 failure; not the same bug." + "reason": "Both are model-specific export/load failures, but not the same concrete code path." }, { "left": "issue:35545", "right": "issue:41720", "accept": false, - "reason": "ONNX export error and auto device-mapping cuda assert are unrelated." + "reason": "ONNX export and device-map cuda assert are unrelated." }, { "left": "issue:45901", "right": "issue:46015", "accept": false, - "reason": "Table QA crash versus outdated RAG docs; one is runtime behavior, the other is documentation." + "reason": "Table QA crash and outdated RAG docs are different issues." }, { "left": "issue:44509", "right": "issue:44512", "accept": false, - "reason": "Both are stale docs after v5 removals, but they reference different removed content and different docs fixes." + "reason": "Both are docs regressions in v5, but they cover different removed features/pages and are not one bug." }, { "left": "issue:35545", "right": "issue:36010", "accept": false, - "reason": "ONNX export issue and GenerationMixin import error are different regressions." + "reason": "ONNX export failure and GenerationMixin import failure are different breakages." }, { "left": "issue:44805", "right": "issue:45823", "accept": false, - "reason": "Both are forward-time errors, but the mask-shape mismatch and device-mismatch during forward are different code paths." + "reason": "Mask shape mismatch and Gemma4 PLE device mismatch are distinct runtime errors." }, { "left": "issue:43638", "right": "issue:44805", "accept": false, - "reason": "Both are IndexError-style reports, but one is ZeRO-3 init/index-0 OOB and the other is a mask/index shape mismatch; not the same bug." + "reason": "Different error signatures and different model/setup contexts." }, { "left": "issue:43596", "right": "issue:44805", "accept": false, - "reason": "Same broad error family, but different underlying symptoms and contexts; not enough to call them duplicates." + "reason": "DeepSpeed ZeRO-3 Bert init error is not the same as the mask-shape mismatch." }, { - "left": "issue:45161", - "right": "issue:44805", + "left": "issue:44805", + "right": "issue:45161", "accept": false, - "reason": "GPT-OSS tensor-parallel issue is unrelated to the mask-shape IndexError." + "reason": "Mask indexing failure and GPT-OSS TP failure are unrelated." }, { "left": "issue:44805", "right": "issue:44918", "accept": false, - "reason": "Different failure modes in different trainers: mask-shape indexing versus unpacking input embeddings." + "reason": "Different models and different failures." }, { "left": "issue:34634", "right": "issue:35545", "accept": false, - "reason": "BarkProcessor voice preset is unrelated to ModernBERT ONNX export." + "reason": "BarkProcessor voice_preset and ModernBERT ONNX export are unrelated." }, { "left": "issue:30277", "right": "issue:43638", "accept": false, - "reason": "Both involve DeepSpeed, but collective mismatch and zero-sized index errors are different bugs." + "reason": "Both involve DeepSpeed, but one is a collective mismatch and the other is a ZeRO-3 index error." }, { "left": "issue:43596", "right": "issue:43638", "accept": true, - "reason": "Same exact ZeRO-3 index-0 out-of-bounds failure path with only model/context details differing." + "reason": "Same DeepSpeed ZeRO-3 index-0-out-of-bounds failure on Bert models; these look like the same underlying bug." }, { "left": "issue:43039", "right": "issue:45161", "accept": false, - "reason": "Liger Kernel cross_entropy invocation and GPT-OSS tensor-parallel failure are unrelated." + "reason": "Liger Kernel cross-entropy being called and GPT-OSS TP not working are different issues." }, { "left": "issue:43975", "right": "issue:44457", "accept": false, - "reason": "DeepSeek detokenization regression is unrelated to LoRA merge/save/load inconsistency." + "reason": "Detokenization regression and LoRA save/load mismatch are not the same bug." }, { "left": "issue:44292", "right": "issue:44457", "accept": false, - "reason": "Different models and failure mechanisms; no shared code path." + "reason": "Qwen NVFP4 runtime failure and LoRA merge/save mismatch are unrelated." }, { "left": "issue:34928", "right": "issue:44805", "accept": false, - "reason": "Activation checkpointing with FSDP is a different distributed-training bug than the mask shape mismatch." + "reason": "Activation checkpointing/FSDP tensor recomputation mismatch is not the same as the mask-shape index error." }, { "left": "issue:24643", "right": "issue:30277", "accept": false, - "reason": "Both mention DeepSpeed, but the training weight-dimension error and collective mismatch are distinct problems." + "reason": "Different DeepSpeed errors with different symptoms and code paths." }, { "left": "issue:45859", "right": "issue:45902", "accept": false, - "reason": "Same model family, but forward CPU-offload incompatibility and uninitialized-memory after materialization are different bugs." + "reason": "Both are Qwen3.5 MoE rotary issues, but one is CPU offload compatibility and the other is uninitialized memory after to_empty materialization." }, { "left": "issue:43638", "right": "issue:45161", "accept": false, - "reason": "ZeRO-3 index-0 OOB is unrelated to GPT-OSS tensor-parallel routing." + "reason": "DeepSpeed Bert index error and GPT-OSS TP failure are unrelated." }, { "left": "issue:34928", "right": "issue:36331", "accept": false, - "reason": "FSDP activation checkpointing issue and CustomTrainer compute_loss signature breakage are unrelated." + "reason": "Checkpointing/FSDP recomputation mismatch and custom trainer signature change are different bugs." }, { "left": "issue:43596", "right": "issue:45161", "accept": false, - "reason": "Different distributed-training failures; no common underlying bug." + "reason": "Bert ZeRO-3 index error is unrelated to GPT-OSS tensor-parallel issues." }, { "left": "issue:43716", "right": "issue:45507", "accept": false, - "reason": "Image preprocessor dtype mismatch is unrelated to GraniteMoEHybrid invalid method calls." + "reason": "Different model families and failure modes." }, { "left": "issue:39290", "right": "issue:45081", "accept": false, - "reason": "Different configs and errors: missing sliding_window_pattern versus Mistral regex patch crashing." + "reason": "Gemma3/vLLM sliding-window attribute error and Mistral tokenizer regex crash are unrelated." }, { "left": "issue:45823", "right": "issue:45874", "accept": false, - "reason": "Device-map mismatch in Gemma4 PLE and reverse embedding lookup are different Gemma4 issues." + "reason": "Device-map forward mismatch and Gemma4 reverse embedding lookup are different problems." }, { "left": "issue:43425", "right": "issue:44292", "accept": false, - "reason": "Torch version incompatibility is not the same as the Qwen-3 NVFP4 runtime error." + "reason": "Torch version incompatibility and Qwen NVFP4 runtime failure are unrelated." }, { "left": "issue:34928", "right": "issue:41720", "accept": false, - "reason": "FSDP checkpointing and Qwen3 auto device-mapping cuda assert are unrelated distributed/runtime bugs." + "reason": "FSDP checkpointing mismatch and cuda assert under auto device mapping are different bugs." }, { "left": "issue:43828", "right": "issue:45161", "accept": false, - "reason": "Autocast dtype mismatch in Phi-tiny-MoE is unrelated to GPT-OSS tensor-parallel failure." + "reason": "Autocast dtype mismatch and GPT-OSS TP failure are not the same code path." }, { "left": "issue:44918", "right": "issue:45161", "accept": false, - "reason": "TRL SFT embedding unpacking and GPT-OSS only-TP failure are different trainer/model issues." + "reason": "Qwen3.5 embedding unpacking with TRL and GPT-OSS TP failure are unrelated." }, { "left": "issue:43716", - "right": "issue:44292", + "right": "issue:45685", "accept": false, - "reason": "Different models and symptoms; not the same underlying bug." + "reason": "Different model/runtime bugs." }, { - "left": "issue:44292", - "right": "issue:44918", + "left": "issue:43828", + "right": "issue:44292", "accept": false, - "reason": "Qwen-3 NVFP4 runtime error and Qwen3.5 input-embedding unpacking are distinct issues." + "reason": "Autocast dtype mismatch and Qwen NVFP4 failure are different issues." }, { "left": "issue:43039", - "right": "issue:43828", + "right": "issue:44918", "accept": false, - "reason": "Liger Kernel cross_entropy routing and Phi autocast dtype mismatch are unrelated." + "reason": "Liger Kernel cross-entropy call and TRL embedding unpacking are different problems." }, { "left": "issue:41762", "right": "issue:43638", - "accept": true, - "reason": "Same ZeRO-3 index-0 out-of-bounds failure; Gemma3-specific and Bert-specific reports look like the same core bug." + "accept": false, + "reason": "Both are ZeRO-3 index errors, but they involve different models and likely different initialization paths." }, { "left": "issue:43827", "right": "issue:44512", "accept": false, - "reason": "Both are docs regressions, but they reference different removed features and likely need separate updates." + "reason": "Both are docs regressions, but they reference different removed features and should not be merged as one duplicate." }, { "left": "issue:43716", "right": "issue:44560", "accept": false, - "reason": "Image dtype mismatch and Qwen3-vl StopIteration are different multimodal failures." + "reason": "Different multimodal model failures and different symptoms." }, { "left": "issue:43828", "right": "issue:45685", "accept": false, - "reason": "Autocast dtype mismatch and mps histogram op not implemented are unrelated." + "reason": "Autocast dtype mismatch and MPS Int histogram error are unrelated." }, { "left": "issue:44918", "right": "issue:45685", "accept": false, - "reason": "TRl trainer embedding unpacking is unrelated to the mps histogram op bug." + "reason": "TRL embedding unpacking and MPS histogram error are unrelated." }, { "left": "issue:43039", "right": "issue:43828", "accept": false, - "reason": "Different model paths and different failure mechanisms." + "reason": "Liger Kernel cross-entropy path and autocast dtype mismatch are different bugs." }, { "left": "issue:30277", "right": "issue:36331", "accept": false, - "reason": "DeepSpeed collective mismatch and trainer loss-signature breakage are not the same issue." + "reason": "DeepSpeed collective mismatch and custom trainer signature mismatch are unrelated." }, { "left": "issue:45237", - "right": "issue:46015", + "right": "issue:45901", "accept": false, - "reason": "Runtime failure on AMD GPUs is unrelated to outdated examples in docs." + "reason": "AMD GPU support issue and table QA crash are unrelated." }, { "left": "issue:34928", "right": "issue:36010", "accept": false, - "reason": "FSDP checkpointing issue and GenerationMixin import error are unrelated." + "reason": "Activation checkpointing/FSDP mismatch and GenerationMixin import failure are unrelated." }, { "left": "issue:43039", "right": "issue:44805", "accept": false, - "reason": "Liger Kernel cross_entropy call and mask/index shape mismatch are different bugs." + "reason": "Liger Kernel cross-entropy call and mask-shape mismatch are different bugs." }, { "left": "issue:41762", "right": "issue:43596", - "accept": true, - "reason": "Same ZeRO-3 index-0 out-of-bounds failure; only the model context differs." + "accept": false, + "reason": "Same broad DeepSpeed/ZeRO-3 area, but not enough evidence they are the same underlying defect." }, { "left": "issue:43901", "right": "issue:44512", "accept": false, - "reason": "TextClassificationPipeline docs and run-command docs are separate documentation discrepancies." + "reason": "Docs mention different deprecated behaviors; not the same issue." }, { "left": "issue:39290", "right": "issue:41720", "accept": false, - "reason": "Different model families and different errors; no shared underlying defect." + "reason": "vLLM sliding-window attribute error and cuda assert under auto device mapping are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "Different subsystems and different failures." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are not the same bug." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross-entropy issue and Qwen3-vl StopIteration are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization bug and RAG docs update are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same AutoTokenizer wrong-token-ID regression; 45920 is the broader multi-model report of the same bug." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading and outdated RAG examples are unrelated." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Checkpoint corruption and NemotronH load failure are different issues." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen NVFP4 runtime failure and RAG docs are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Gemma3/vLLM sliding-window attribute error and Qwen3-MoE sliding_window issue are different model/config bugs." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Similar IndexError wording, but different shapes, models, and contexts; not clearly the same bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the soft-edge judgments are appropriately conservative. The ZeRO-3 IndexError cluster is well supported by the issue titles, and the docs/model-specific rejections do not overclaim duplicates." + "feedback": "The cluster summary is grounded in the packet, and the two accepted duplicate pockets are well supported by the titles: DeepSpeed ZeRO-3 Bert index errors (43596/43638) and the AutoTokenizer v5 wrong-token-ID regression (45812/45920). The rejected links are generally conservative and do not overstate evidence, and the canonical issue choice of 45920 is justified as the broader umbrella report." }, "evaluator_used": true, "retried": false, @@ -496,8 +556,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -560,14 +620,14 @@ "issue:45362|issue:45464", "issue:44292|issue:45901", "issue:39290|issue:45823", - "issue:43404|issue:43866", "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:41093|issue:41762", "issue:45800|issue:45823", + "issue:41093|issue:41762", "issue:43572|issue:45070", "issue:43296|issue:43366", "issue:43541|issue:45161", @@ -590,269 +650,272 @@ "issue:43404|issue:45005", "issue:44530|issue:45005" ], - "prepared_review_unit_hash": "18750e6fbcac204e9957863c0cda20e9cbd8ef3f56634188682e3037990b3d19", + "prepared_review_unit_hash": "81f617f1c961dc0df36ede00d0176fd04ff5c68eb3cfbc116646d436aeff94a4", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 38, - "serialized_chars": 38797, - "estimated_input_tokens": 9700, - "estimated_eval_tokens": 19656 + "serialized_chars": 38793, + "estimated_input_tokens": 9699, + "estimated_eval_tokens": 19654 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is not a real duplicate cluster: the items span many unrelated model-loading, dtype, quantization, zero3, and API regressions. The soft links are mostly driven by generic wording rather than the same underlying bug.", - "confidence": 0.97, - "canonical_issue_reason": "No canonical issue stands out; the issues are too heterogeneous to represent one duplicate set.", + "summary": "Mostly a collection of distinct model/runtime regression reports rather than one duplicate cluster. I only found one plausible duplicate pair around Siglip2 output quality; the rest are too model- or path-specific to merge confidently.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45020 is the broadest and most central report in the set, capturing the recurring 'recent transformers versions break model loading/behavior' theme better than the narrower model-specific bugs.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a strong representative because there is no coherent underlying duplicate cluster.", + "best_issue_reason": "issue:45020 is the best cluster representative because it is the most general compatibility/regression issue and most likely to subsume related follow-up reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43854", "right": "issue:43866", "accept": false, - "reason": "Different models and different failure modes: unit-test load failure vs corrupted checkpoint." + "reason": "Different models and failures: GLM-4.7-Flash loading in tests vs Ovis2 checkpoint corruption." }, { "left": "issue:43792", "right": "issue:45901", "accept": false, - "reason": "Whisper loading/runtime failure and table-QA crash are unrelated bugs." + "reason": "Whisper inference failure and table-question-answering crash are unrelated code paths." }, { "left": "issue:43425", "right": "issue:45070", "accept": false, - "reason": "Torch version incompatibility and a pydantic PretrainedConfig regression are different issues." + "reason": "Torch 2.10 compatibility and a PretrainedConfig pydantic field regression are different bugs." }, { "left": "issue:43596", "right": "issue:45800", "accept": false, - "reason": "Deepspeed zero3 index error and a torch/transformers version incompatibility are not the same bug." + "reason": "DeepSpeed ZeRO-3/BertModel index error is not the same as a torch 2.4.1 vs transformers 5.8 incompatibility." }, { "left": "issue:43975", "right": "issue:45901", "accept": false, - "reason": "Detokenization bug and table-question-answering crash are unrelated." + "reason": "Incorrect detokenization in v5 and table QA crashes are unrelated." }, { "left": "issue:43792", "right": "issue:45070", "accept": false, - "reason": "Whisper can't run and PretrainedConfig pydantic breakage are different failure paths." + "reason": "Different failures with no shared concrete code path." }, { "left": "issue:45362", "right": "issue:45464", "accept": false, - "reason": "Same model family, but chat crash vs streaming API failure are different code paths and symptoms." + "reason": "Both mention Qwen3.5, but one is a chat crash and the other is a streaming inference API failure on a different model size/path." }, { "left": "issue:44292", "right": "issue:45901", "accept": false, - "reason": "init_empty_weights TypeError is unrelated to a table-QA crash." + "reason": "Unrelated model/runtime error and table QA crash." }, { "left": "issue:39290", "right": "issue:45823", "accept": false, - "reason": "Different models and different problems: missing sliding_window_pattern vs device mismatch." + "reason": "Gemma3 sliding-window attribute error and Gemma4 device mismatch are different regressions." }, { - "left": "issue:43404", - "right": "issue:43866", + "left": "issue:45464", + "right": "issue:45685", "accept": false, - "reason": "Tied lm_head weights bug and a corrupted checkpoint are not the same underlying issue." + "reason": "Qwen3.5 streaming failure and an MPS histogram kernel limitation are unrelated." }, { - "left": "issue:45464", - "right": "issue:45685", + "left": "issue:43404", + "right": "issue:43866", "accept": false, - "reason": "Streaming inference failure and MPS histogram int error are unrelated." + "reason": "Weight-tying bug in Mistral3 is unrelated to a corrupted Ovis2 checkpoint." }, { "left": "issue:43541", "right": "issue:45685", "accept": false, - "reason": "Grouped_mm tracing error and MPS histogram int error are different runtime issues." + "reason": "Grouped_mm tracing failure in Mixtral is not the same as the MPS Int histogram error." }, { "left": "issue:43541", "right": "issue:43828", "accept": false, - "reason": "Both are dtype-related at a high level, but they affect different models and code paths." + "reason": "Both involve runtime numeric issues, but they are different models and different failure mechanisms." }, { "left": "issue:44841", "right": "issue:45084", "accept": false, - "reason": "Processor load failure and template compilation error are unrelated." + "reason": "Voxtral processor failure and a template-node compile TypeError are unrelated." }, { "left": "issue:36010", "right": "issue:39290", "accept": false, - "reason": "ImportError for GenerationMixin and Gemma3 sliding_window attr missing are unrelated." + "reason": "ImportError for GenerationMixin and a Gemma3 config attribute error are different bugs." }, { - "left": "issue:41093", - "right": "issue:41762", + "left": "issue:45800", + "right": "issue:45823", "accept": false, - "reason": "Both mention shape/indexing errors, but they involve different tensors, models, and failure points." + "reason": "Generic torch/transformers incompatibility and Gemma4 device placement mismatch are not the same issue." }, { - "left": "issue:45800", - "right": "issue:45823", + "left": "issue:41093", + "right": "issue:41762", "accept": false, - "reason": "Torch compatibility and Gemma4 device mismatch are different bugs." + "reason": "Both are index-related, but one is a mask shape mismatch and the other is a ZeRO-3 load-time out-of-bounds error." }, { "left": "issue:43572", "right": "issue:45070", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is unrelated to a pydantic field regression." + "reason": "Missing pad_token_idx in StableLmConfig and a pydantic field regression are different breakages." }, { "left": "issue:43296", "right": "issue:43366", "accept": false, - "reason": "PaddleOCR-VL loading failure and GGUF support request are not duplicates." + "reason": "PaddleOCR-VL load failure in vLLM and GGUF gpt-oss support are different feature/issues." }, { "left": "issue:43541", "right": "issue:45161", "accept": false, - "reason": "Both are MoE/parallelism complaints, but they are different concrete failures." + "reason": "Mixtral torch-dynamo tracing error and GPT-OSS tensor parallelism failure are different code paths." }, { "left": "issue:44387", "right": "issue:45005", "accept": false, - "reason": "Int4 OOM regression and tied-weights translation bug are unrelated." + "reason": "Int4 OOM regression and tied-weights translation-model issue are unrelated." }, { "left": "issue:43828", "right": "issue:45464", "accept": false, - "reason": "Autocast dtype mismatch and streaming API failure are different issues." + "reason": "Phi-tiny-MoE autocast dtype mismatch and Qwen3.5 streaming API failure are different bugs." }, { "left": "issue:44291", "right": "issue:45070", "accept": false, - "reason": "init_empty_weights argument breakage is not the same as the PretrainedConfig pydantic issue." + "reason": "init_empty_weights/_is_hf_initialized argument handling is unrelated to the pydantic model breakage." }, { "left": "issue:43366", "right": "issue:45464", "accept": false, - "reason": "GGUF model support and Qwen3.5 streaming failure are unrelated." + "reason": "GGUF gpt-oss support and Qwen3.5 chat streaming failure are unrelated." }, { "left": "issue:45020", "right": "issue:45800", "accept": false, - "reason": "Remote-code regression and torch/transformers incompatibility are different problems." + "reason": "Remote_code model breakage and a torch version incompatibility are too broad and distinct to treat as one bug." }, { "left": "issue:43531", "right": "issue:45070", "accept": false, - "reason": "Qwen3-MoE sliding_window issue and pydantic model regression are unrelated." + "reason": "Qwen3-MoE sliding_window behavior and a pydantic field regression are different issues." }, { "left": "issue:43541", "right": "issue:44589", "accept": false, - "reason": "Grouped_mm tracing failure and missing Float8 storage type are unrelated runtime errors." + "reason": "Grouped_mm tracing failure and missing Float8 storage object are unrelated runtime errors." }, { "left": "issue:41093", "right": "issue:43596", "accept": false, - "reason": "Both are zero3-ish indexing failures, but the symptoms and affected paths differ too much to merge." + "reason": "Both are index errors, but one is a mask-length mismatch and the other is a zero-size tensor access during ZeRO-3 init." }, { "left": "issue:43054", "right": "issue:43994", - "accept": false, - "reason": "Both concern SigLIP2 behavior, but one is a quality regression and the other is a general output failure; not clearly the same bug." + "accept": true, + "reason": "Both point to Siglip2 producing bad/nonsensical text-embedding-driven results, likely the same underlying model/processor issue." }, { "left": "issue:45542", "right": "issue:45800", "accept": false, - "reason": "TensorBoard-without-TF backend error and torch/transformers incompatibility are unrelated." + "reason": "TensorBoard-without-TensorFlow backend error and torch/transformers incompatibility are unrelated." }, { "left": "issue:45759", "right": "issue:45902", "accept": false, - "reason": "Composite config unwrapping and uninitialized inv_freq memory are distinct model-initialization bugs." + "reason": "Composite Qwen config unwrapping and uninitialized rotary embedding memory are different initialization bugs." }, { "left": "issue:39186", "right": "issue:44918", "accept": false, - "reason": "FSDP weight-dimension error and Qwen3.5 embedding unpacking failure are different issues." + "reason": "FSDP 2-D weight assertion and TRL embedding unpacking failure are different problems." }, { "left": "issue:43716", "right": "issue:44918", "accept": false, - "reason": "Image-preprocessor dtype mismatch and trl SFT embedding unpacking failure are unrelated." + "reason": "Mistral-3 image-preprocessor dtype mismatch and Qwen3.5 TRL unpacking failure are unrelated." }, { "left": "issue:40990", "right": "issue:44960", "accept": false, - "reason": "High perplexity on gpt-oss-20b and a GLM5 issue are not duplicates." + "reason": "GPT-OSS perplexity quality issue and GLM5 are not the same bug." }, { "left": "issue:45632", "right": "issue:45698", "accept": false, - "reason": "Both involve custom-module loading, but one is a leaf-directory cache collision and the other is wrong module selection after save_pretrained; not the same concrete bug." + "reason": "Both involve custom-module loading, but one is a trust_remote_code cache path collision and the other is a wrong module loaded after save_pretrained; not enough evidence of the same concrete bug." }, { "left": "issue:43645", "right": "issue:45399", "accept": false, - "reason": "Jupyter custom-model initialization and flash-attn2 fallback checks are unrelated." + "reason": "Jupyter custom-model initialization regression and flash-attn2 fallback blocking are unrelated." }, { "left": "issue:43404", "right": "issue:45005", "accept": false, - "reason": "Tied weights in Mistral3 and tied weights on translation models are different model-specific regressions." + "reason": "Mistral3 lm_head tying and translation-model weight tying are similar in theme but not the same concrete bug." }, { "left": "issue:44530", "right": "issue:45005", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and tied-weights translation regression are unrelated." + "reason": "PagedAttention linear_attention crash and translation-model tied weights are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is conservative and stays grounded in the packet: most pairs are plausibly distinct based on titles, and the single accepted Siglip2 pair is the only one with a reasonable shared-bug reading. The choice of issue:45020 as a representative is subjective but defensible given its broad compatibility/regression framing." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -864,8 +927,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -932,8 +995,8 @@ "issue:43828|issue:44928", "issue:36331|issue:41762", "issue:41628|issue:45800", - "issue:43950|issue:45902", "issue:43872|issue:44589", + "issue:43950|issue:45902", "issue:43866|issue:44403", "issue:43054|issue:45701", "issue:41553|issue:43883", @@ -951,235 +1014,241 @@ "issue:41762|issue:42915", "issue:45685|issue:45823" ], - "prepared_review_unit_hash": "b8c5b6ead19c2dd09d8a0d94fcb83396b9c4afa9d49d4986c19373e60db8b03d", + "prepared_review_unit_hash": "10c469a24391574294c85ab89264bc5068c59504979715b6ecb21e8fd2154ca9", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36383, - "estimated_input_tokens": 9096, - "estimated_eval_tokens": 18448 + "serialized_chars": 36380, + "estimated_input_tokens": 9095, + "estimated_eval_tokens": 18446 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a mixed bag of unrelated Transformers bug reports. The only clearly duplicate family is the non-persistent-buffer regression in `from_pretrained`/`to_empty`, with a smaller SigLIP2 implementation-discrepancy family. No PRs are present.", - "confidence": 0.78, - "canonical_issue_reason": "Issue 43950 is the clearest root-cause report for the non-persistent-buffer regression; 44534 and 45902 look like later/specific manifestations of the same bug.", + "summary": "Mostly a mixed similarity cluster. The safest duplicate chain is the v5 non-persistent-buffer regression around 43950 and 44534; 45902 is related but not strong enough to confirm from the packet alone. The SigLIP2 pair 38175/43054 also looks related, but the evidence is too thin to call it a confirmed duplicate. The Qwen3.5 3D position_ids pair (44910/44928) is the other clear duplicate-like edge.", + "confidence": 0.72, + "canonical_issue_reason": "43950 is still the clearest root-cause report for the non-persistent-buffer v5 regression, with 44534 as the strongest corroborating duplicate-like report.", "canonical_pr_reason": null, - "best_issue_reason": "43950 is the best dedupe anchor because it states the underlying regression directly and broadly, making it the most useful canonical issue for that subcluster.", + "best_issue_reason": "43950 is the best representative issue because it states the regression most concretely and anchors the strongest duplicate chain in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42915", "right": "issue:45399", "accept": false, - "reason": "Different failures: FP8 config loading vs flash-attn2 fallback gating." + "reason": "Different failures: FP8/Qwen3 MoE loading vs flash-attn fallback gating; same area, not the same bug." }, { "left": "issue:43873", "right": "issue:45005", "accept": false, - "reason": "Quantization/offloading behavior and tied-weights handling are unrelated bugs." + "reason": "Quantization/offloading memory behavior vs tied-weights translation-model bug; no shared concrete code path." }, { "left": "issue:44360", "right": "issue:44512", "accept": false, - "reason": "Completely different areas: DSA indexer implementation vs docs text." + "reason": "DSA indexer math issue vs a docs typo; unrelated." }, { "left": "issue:44863", "right": "issue:45005", "accept": false, - "reason": "Checkpoint loading for NemotronH is unrelated to translation-model tied-weights regression." + "reason": "NemotronH checkpoint-loading failure vs translation tied-weights regression; different model paths." }, { "left": "issue:44918", "right": "issue:45507", "accept": false, - "reason": "Different model paths and symptoms: TRL unpacking vs GraniteMoEHybrid invalid method." + "reason": "Qwen3.5 embedding unpacking in TRL vs GraniteMoE invalid method call; not the same defect." }, { "left": "issue:44492", "right": "issue:44512", "accept": false, - "reason": "A docs typo and a removed command mention are not the same issue." + "reason": "Cache-strategy typo vs removed-command documentation; editorial, not duplicate bugs." }, { "left": "issue:43975", "right": "issue:44292", "accept": false, - "reason": "Different models and different runtime failures; no shared concrete code-path." + "reason": "Detokenization regression vs Qwen-3 NVFP4 load/run failure; different symptoms and paths." }, { "left": "issue:44292", "right": "issue:45507", "accept": false, - "reason": "Distinct model-specific errors with no indication of the same root cause." + "reason": "Separate model-loading/runtime failures with no clear shared root cause." }, { "left": "issue:43950", "right": "issue:44534", "accept": true, - "reason": "Same non-persistent-buffer corruption regression, reported with slightly different wording." + "reason": "Same non-persistent-buffer v5 regression; 44534 matches the junk/corruption symptom reported in 43950." }, { "left": "issue:43643", "right": "issue:43950", "accept": false, - "reason": "Remote-code missing-fields bug is a different failure mode from non-persistent buffer corruption." + "reason": "Remote-code config-field loss vs non-persistent-buffer corruption; different bugs." }, { "left": "issue:39692", "right": "issue:43054", "accept": false, - "reason": "Both involve SigLIP2, but one is a docs example/processor mismatch and the other is an embedding-quality regression." + "reason": "Docs example/quantization failure vs SigLIP2 embedding-quality regression; too different." }, { "left": "issue:43828", "right": "issue:44928", "accept": false, - "reason": "Different model families and different dtype/packing failures." + "reason": "Autocast dtype mismatch on Phi-tiny-MoE vs Qwen3.5 3D position_ids/SDPA instability; not the same issue." }, { "left": "issue:36331", "right": "issue:41762", "accept": false, - "reason": "Trainer API mismatch and ZeRO-3 loading error are unrelated." + "reason": "CustomTrainer API change vs Gemma3 ZeRO-3 loading crash; unrelated." }, { "left": "issue:41628", "right": "issue:45800", "accept": false, - "reason": "Import/export issues and torch-version incompatibility are too broad to merge." - }, - { - "left": "issue:43950", - "right": "issue:45902", - "accept": true, - "reason": "Same underlying non-persistent-buffer initialization bug; 45902 is a concrete manifestation of the same regression." + "reason": "Import/export issue for AutoImageProcessor vs torch/transformers version incompatibility; not a duplicate." }, { "left": "issue:43872", "right": "issue:44589", "accept": false, - "reason": "Bitsandbytes parameter incompatibility and missing storage-object error are different breakages." + "reason": "bitsandbytes kwarg incompatibility vs missing Float8 storage; different incompatibilities." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Related non-persistent-buffer regression, but the packet does not firmly establish the same concrete bug/root cause; safer as a candidate duplicate than a confirmed one." }, { "left": "issue:43866", "right": "issue:44403", "accept": false, - "reason": "Checkpoint corruption and loading noise are unrelated." + "reason": "Checkpoint corruption vs generic loading noise; no shared defect." }, { "left": "issue:43054", "right": "issue:45701", "accept": false, - "reason": "SigLIP2 embedding quality vs tokenizer-version drift are not the same bug." + "reason": "SigLIP2 embedding-quality regression vs tokenizer/version behavior change; different subsystems." }, { "left": "issue:41553", "right": "issue:43883", "accept": false, - "reason": "Bad AutoTokenizer error message and missing all_tied_weights_keys are separate code paths." + "reason": "Bad tokenizer error message vs missing tied-weights attribute; unrelated." }, { "left": "issue:38175", "right": "issue:43054", - "accept": true, - "reason": "Both describe the same SigLIP2 text/score degeneration family." + "accept": false, + "reason": "Both mention SigLIP2 output quality problems, but the packet does not clearly show the same bug/root cause; too weak to confirm as duplicates." }, { "left": "issue:44910", "right": "issue:45859", "accept": false, - "reason": "FlashAttention 3D position_ids crash and CPU-offload incompatibility are distinct." + "reason": "Qwen3.5 flash-attn/packed-sequence crash vs CPU-offload incompatibility in rotary embedding; different code paths." }, { "left": "issue:45663", "right": "issue:45902", "accept": false, - "reason": "Gemma4 FSDP2 key error is unrelated to non-persistent-buffer materialization." + "reason": "Gemma4 FSDP2 KeyError vs uninitialized rotary inv_freq after to_empty(); different models and failures." }, { "left": "issue:41553", "right": "issue:43881", "accept": false, - "reason": "Different model-loading problems for different models." + "reason": "Voxtral tokenizer error message vs glm-4v loading failure; unrelated model-loading problems." }, { "left": "issue:42947", "right": "issue:44387", "accept": false, - "reason": "Gradient checkpointing/LoRA and int4 CUDA memory regression are different issues." + "reason": "Gradient-checkpointing/LoRA issue vs int4 quantization reserved-memory OOM; different root causes." }, { "left": "issue:43054", "right": "issue:43493", - "accept": true, - "reason": "Same SigLIP2 implementation discrepancy; one is a symptom, the other compares HF and JAX implementations." + "accept": false, + "reason": "Very related SigLIP2 reports, but one is a broad HF-vs-JAX discrepancy and the other is a symptom report; not enough to call them the same issue." }, { "left": "issue:44403", "right": "issue:44451", "accept": false, - "reason": "Different model load failures with no evidence of one shared root cause." + "reason": "General loading noise vs a specific model load failure; not the same bug." }, { "left": "issue:43643", "right": "issue:45020", "accept": false, - "reason": "Remote-code field loss is more specific than a broad compatibility regression; not enough to merge." + "reason": "Specific AutoConfig remote_code field loss vs broad remote_code breakage report; too generic to merge confidently." }, { "left": "issue:43975", "right": "issue:45237", "accept": false, - "reason": "Different models and unrelated failure modes." + "reason": "Detokenization bug vs AMD GPU support/loading problem; unrelated." }, { "left": "issue:44403", "right": "issue:44488", "accept": false, - "reason": "Separate loading reports for different models; too weak to treat as duplicates." + "reason": "Generic loading noise vs a specific model load failure; distinct reports." }, { "left": "issue:42947", "right": "issue:43856", "accept": false, - "reason": "LoRA checkpointing inefficiency and Qwen3 MoE memory use are different performance issues." + "reason": "LoRA checkpointing issue vs Qwen3 MoE memory usage concern; different problems." }, { "left": "issue:41762", "right": "issue:42915", "accept": false, - "reason": "Gemma3 ZeRO-3 loading error and Qwen3Moe FP8 failure are unrelated." + "reason": "Gemma3 ZeRO-3 loading crash vs Qwen3 MoE FP8/flash-attn loading issue; unrelated." }, { "left": "issue:45685", "right": "issue:45823", "accept": false, - "reason": "MPS histogram kernel error and Gemma4 device mismatch are different model/runtime problems." + "reason": "MPS histogram op missing in MoE vs Gemma4 device-mismatch forward bug; different defects." + }, + { + "left": "issue:44910", + "right": "issue:44928", + "accept": true, + "reason": "Same Qwen3.5 3D position_ids handling bug; one crashes flash-attn, the other triggers SDPA/BF16 instability." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The cluster summary is grounded in the packet: the non-persistent-buffer issues form a coherent duplicate family, the SigLIP2 items plausibly form a smaller related family, and the rest are conservatively treated as unrelated. The soft-edge decisions are mostly cautious and supported by the titles/body metadata. No PR pair concerns apply because there are no PRs." + "accept": false, + "feedback": "The summary is mostly grounded, but a couple of the accepted soft-edge pairs are a bit too strong for the evidence in the packet. In particular, 43950\u219445902 and 38175\u219443054 look related, but the titles alone do not clearly establish the same bug/root cause; they are safer as candidate duplicates than confirmed ones. Because the task asks for conservative soft-edge verdicts, I\u2019d reject this as slightly overconfident." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -1192,8 +1261,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -1269,9 +1338,9 @@ "issue:41553|issue:44991", "issue:43940|issue:45313", "issue:36331|issue:41093", - "issue:43854|issue:44863", "issue:42947|issue:45923", "issue:44928|issue:45588", + "issue:43854|issue:44863", "issue:43257|issue:45305", "issue:43872|issue:45538", "issue:41093|issue:44805", @@ -1280,232 +1349,232 @@ "issue:44530|issue:44863", "issue:24643|issue:30064" ], - "prepared_review_unit_hash": "8f216b4fdd77fadfe3c5de32d3fa9b98f201c074ad79e3a37834c9444f6cbff5", + "prepared_review_unit_hash": "4be4c4b508b5552986984d17dad15c6780305489344efa22e2b1b7abdaacb436", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37613, - "estimated_input_tokens": 9404, - "estimated_eval_tokens": 19064 + "serialized_chars": 37611, + "estimated_input_tokens": 9403, + "estimated_eval_tokens": 19062 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a grab-bag of unrelated Transformers bug reports, with a few small duplicate pockets around docs cleanup, `_is_hf_initialized`/`init_empty_weights`, mask-shape IndexErrors, and a Qwen video regression. No single issue/PR cleanly represents the whole cluster.", - "confidence": 0.71, - "canonical_issue_reason": null, + "summary": "Mostly heterogeneous cluster. The clearest duplicate pairs are the v5 docs pipeline-removal reports and the init_empty_weights/_is_hf_initialized loading regression; there is also a plausible Qwen3-VL video-input regression pair, but the rest look unrelated by model, subsystem, or failure mode.", + "confidence": 0.83, + "canonical_issue_reason": "No single cluster-wide canonical stands out, but if one issue must anchor triage, 44291 is the strongest because it states the exact root cause/version boundary and matches the clearest duplicate counterpart (43872).", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the broadest umbrella for the recurring recent-version model-loading regressions, especially remote_code-related breakage, so it is the best representative issue even though the cluster is not one duplicate set.", + "best_issue_reason": "44291 is the most precise, actionable issue in the set and best represents the concrete loading-regression duplicate pair; the other items are too mixed to serve as a better global anchor.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43782", "right": "issue:44960", "accept": false, - "reason": "Different models and failures: Qwen3VL weight_only load error vs GLM5 loading issue." + "reason": "Different models and failures: Qwen3-VL weight_only loading vs a GLM5 unit-test loading problem." }, { "left": "issue:43881", "right": "issue:43883", "accept": false, - "reason": "Related theme, but different concrete bugs on different model classes." + "reason": "Different code paths: GLM-4V loading failure vs missing all_tied_weights_keys on Molmo." }, { "left": "issue:43824", "right": "issue:43872", "accept": false, - "reason": "Import/export name issue vs bitsandbytes `_is_hf_initialized` constructor incompatibility." + "reason": "ImportError for a missing Qwen2_5_VL class vs a bitsandbytes init TypeError; unrelated." }, { "left": "issue:44991", "right": "issue:45399", "accept": false, - "reason": "Tokenizer-loading regression vs flash-attn fallback gating; not the same bug." + "reason": "Tokenizer loading regression vs flash-attn fallback gating; not the same bug." }, { "left": "issue:45020", "right": "issue:45823", "accept": false, - "reason": "Remote_code regression and device_map auto forward mismatch are different code-path failures." + "reason": "Generic remote_code breakage vs Gemma4 device-map forward mismatch; different code paths." }, { "left": "issue:42491", "right": "issue:44368", "accept": false, - "reason": "LoRA compatibility/load failure vs a tie_word_embeddings warning; not the same issue." + "reason": "Both mention Qwen/LoRA, but one is a model-compatibility regression and the other is just a tie_word_embeddings warning." }, { "left": "issue:44661", "right": "issue:45020", "accept": false, - "reason": "Tokenizer mapping edge case vs general remote_code regression." + "reason": "add-new-model-like handling of TOKENIZER_MAPPING_NAMES is a different failure from remote_code model loading regressions." }, { "left": "issue:43828", "right": "issue:45823", "accept": false, - "reason": "Autocast dtype mismatch is not the same as device placement mismatch." + "reason": "Autocast dtype mismatch and device placement mismatch are distinct runtime problems." }, { "left": "issue:45081", "right": "issue:45399", "accept": false, - "reason": "Mistral regex patch crash vs flash-attn fallback logic." + "reason": "Mistral regex patch crash vs flash-attn fallback selection; unrelated subsystems." }, { "left": "issue:45020", "right": "issue:45356", "accept": false, - "reason": "General loading regression vs Kimi tokenizer codec-handling regression." + "reason": "Remote_code loading regressions vs Kimi tokenizer codec handling; different underlying bugs." }, { "left": "issue:43827", "right": "issue:44509", "accept": true, - "reason": "Both report the same docs cleanup: removing deprecated v5 pipeline tasks from summarization/translation docs." + "reason": "Same docs regression: v5 removed pipeline tasks, but the docs still reference summarization/translation/text2text-generation pipeline usage." }, { "left": "issue:43742", "right": "issue:45399", "accept": false, - "reason": "MobileLLM load failure and flash-attn fallback blocking are unrelated." + "reason": "MobileLLM key error vs flash-attn installation/fallback logic; not the same issue." }, { "left": "issue:43404", "right": "issue:43950", "accept": false, - "reason": "Weight-tying bug in Mistral3 vs non-persistent buffer corruption during from_pretrained." + "reason": "Untied lm_head in Mistral3 and non-persistent buffer corruption are different model/runtime bugs." }, { "left": "issue:42915", "right": "issue:45020", "accept": false, - "reason": "Qwen3 MoE ZeRO-3 weight conversion problem is different from the remote_code regression." + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is a different concrete failure than generic remote_code breakage." }, { "left": "issue:43638", "right": "issue:45823", "accept": false, - "reason": "Zero-3 Bert index error and Gemma4 device mismatch do not share the same underlying bug." + "reason": "ZeRO-3 index-out-of-bounds on a non-pretrained Bert model vs Gemma4 PLE device mismatch; unrelated." }, { "left": "issue:43257", "right": "issue:43866", "accept": false, - "reason": "Qwen3 MoE weight conversion and Ovis2 checkpoint corruption are different failures." + "reason": "Qwen3 MOE weight conversion bug vs Ovis2 checkpoint corruption; different models and symptoms." }, { "left": "issue:43792", "right": "issue:44220", "accept": false, - "reason": "Too little evidence that the Whisper runtime failure and `_torch_extract_fbank_features()` report are the same concrete bug." + "reason": "Whisper startup failure and _torch_extract_fbank_features are related at a very high level, but the titles do not show the same concrete bug or code path." }, { "left": "issue:43872", "right": "issue:44291", "accept": true, - "reason": "Same `_is_hf_initialized`/`init_empty_weights` incompatibility; the titles describe the same TypeError from the same kwarg path." + "reason": "Same init_empty_weights / _is_hf_initialized incompatibility; 44291 spells out the exact regression that 43872 reports more generally." }, { "left": "issue:42915", "right": "issue:43872", "accept": false, - "reason": "DeepSpeed weight-loading issue vs bitsandbytes kwarg incompatibility." + "reason": "DeepSpeed Qwen3 MOE weight-loading bug vs bitsandbytes init error; different failure modes." }, { "left": "issue:41553", "right": "issue:44991", "accept": false, - "reason": "Bad error message for Voxtral loading is unrelated to est-roberta tokenizer loading failure." + "reason": "Bad AutoTokenizer error messaging for Voxtral vs tokenizer loading failure for EMBEDDIA/est-roberta." }, { "left": "issue:43940", "right": "issue:45313", "accept": false, - "reason": "Both are ZeRO-3 weight-loading issues, but they affect different Qwen variants and appear to be distinct bugs." + "reason": "Both are ZeRO-3 weight-loading reports, but they target different Qwen variants and the titles do not indicate the same concrete root cause." }, { "left": "issue:36331", "right": "issue:41093", "accept": false, - "reason": "Trainer API mismatch vs mask/tensor shape IndexError." - }, - { - "left": "issue:43854", - "right": "issue:44863", - "accept": false, - "reason": "Different models and failures; no sign of the same underlying defect." + "reason": "Trainer compute_loss API mismatch vs a mask/tensor shape IndexError; unrelated." }, { "left": "issue:42947", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing effectiveness bug vs missing capability flag on a remote-code variant." + "reason": "Gradient checkpointing ineffective with PEFT LoRA and missing supports_gradient_checkpointing are related broadly, but not the same bug." }, { "left": "issue:44928", "right": "issue:45588", "accept": false, - "reason": "RLHF NaN from 3D position_ids/BF16 collapse is a different issue from the flash_attention s_aux=None crash." + "reason": "RLHF NaNs from 3D position_ids/SDPA fallback vs flash_attention crash on s_aux=None; different mechanisms." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "GLM-4.7 loading in tests vs NemotronH checkpoint loading; unrelated model-specific issues." }, { "left": "issue:43257", "right": "issue:45305", "accept": false, - "reason": "Weight conversion under accelerate+deepspeed vs gradient averaging with GAS are separate training bugs." + "reason": "MOE weight conversion during loading vs gradient averaging under GAS with DeepSpeed; different training bugs." }, { "left": "issue:43872", "right": "issue:45538", "accept": false, - "reason": "bitsandbytes init kwarg issue is unrelated to CLIPTokenizer's model_max_length constant." + "reason": "bitsandbytes init incompatibility vs CLIPTokenizer max_length default; unrelated." }, { "left": "issue:41093", "right": "issue:44805", - "accept": true, - "reason": "Same mask-vs-indexed-tensor shape mismatch IndexError; the later issue is the same underlying error class with placeholders." + "accept": false, + "reason": "The error shape is similar, but the titles do not establish the same root cause or model path; likely separate mask-shape bugs." }, { "left": "issue:44479", "right": "issue:44560", "accept": true, - "reason": "Both describe the same v5.3.0 Qwen video-input regression; 44560 looks like a specific instance of the broader 44479 bug." + "reason": "Both point to the Qwen3-VL video-input regression in 5.3.0; 44560 looks like the embedding-model manifestation of the same StopIteration failure." }, { "left": "issue:41553", "right": "issue:44488", "accept": false, - "reason": "Voxtral tokenizer error-message issue is unrelated to loading cjvt/sleng-bert." + "reason": "Voxtral tokenizer error messaging vs sleng-bert loading failure; different artifacts and failures." }, { "left": "issue:44530", "right": "issue:44863", "accept": false, - "reason": "PagedAttentionCache linear_attention crash and NemotronH checkpoint loading failure are different bugs." + "reason": "PagedAttentionCache linear_attention crash vs NemotronH checkpoint loading; different subsystems and bugs." }, { "left": "issue:24643", "right": "issue:30064", "accept": false, - "reason": "DeepSpeed 'weight must be 2-D' training error and void segmentation map processing failure are unrelated." + "reason": "DeepSpeed training weight shape error vs image processor void-segmentation-map failure; unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is conservative and grounded in the packet. The accepted soft-edge pairs are reasonably supported by the issue titles/bodies: docs cleanup, `_is_hf_initialized`/`init_empty_weights`, the mask-shape IndexError, and the Qwen video regression. The rejected pairs are phrased cautiously and do not overstate equivalence. The choice of issue:45020 as a broad representative is defensible, and the note that the cluster is not a single duplicate set is accurate." + "feedback": "The summary is grounded in the packet and stays conservative. The accepted soft-edge pairs are plausibly the same underlying bug family, and the rejected pairs are generally justified by distinct model/subsystem/failure-mode differences. No overstatement stands out." }, "evaluator_used": true, "retried": false, @@ -1521,8 +1590,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -1604,24 +1673,24 @@ "issue:42491|issue:43720", "issue:41720|issue:42915", "issue:44960|issue:44977", - "issue:45313|issue:45357", "issue:44451|issue:44488", + "issue:45313|issue:45357", "issue:42915|issue:44756" ], - "prepared_review_unit_hash": "48a5eab2114ee5d42f5cd9a0ce5be21532eff54c6f6ac58871037d42dd93bc2d", + "prepared_review_unit_hash": "e6bfe591091595960c9618c14527c1547186951444b0889dc1e2fd75f6668dac", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36660, + "serialized_chars": 36659, "estimated_input_tokens": 9165, "estimated_eval_tokens": 18586 }, @@ -1630,210 +1699,216 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of unrelated Transformers regressions, with only a few clear duplicate pairs. The strongest matches are the non-persistent buffer regression, the custom-model initialization regression, and a SigLIP2 inference-output regression.", - "confidence": 0.83, - "canonical_issue_reason": null, + "summary": "The cluster is mostly heterogeneous, but a few tight duplicate subgroups stand out: non-persistent buffer regressions, SigLIP2 output corruption, Qwen3.5 flash-attention failures, and a couple of model-loading regressions.", + "confidence": 0.76, + "canonical_issue_reason": "issue:43644 is the clearest and earliest report of the exact non-persistent-buffer regression, and it has a near-exact duplicate in issue:44534.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43644 is the cleanest representative: it has a precise, self-contained regression description and an exact duplicate in 44534.", + "best_issue_reason": "issue:43644 is the best overall issue anchor here: specific, regression-focused, and the strongest exact-match duplicate candidate in the set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43366", "right": "issue:43828", "accept": false, - "reason": "Different models and failure modes: GGUF/gpt-oss support vs Phi-tiny-MoE autocast dtype mismatch." + "reason": "Different models and failure modes; no shared concrete code path." }, { "left": "issue:45325", "right": "issue:45902", "accept": false, - "reason": "Both are Qwen-family, but one is a VL rope-scaling bug and the other is a Qwen3 MoE meta-materialization bug." + "reason": "Different bug classes: rope-position scaling vs uninitialized rotary embedding memory." }, { "left": "issue:45663", "right": "issue:45823", "accept": false, - "reason": "Same model family, but distinct code paths: FSDP2 shared_kv_states KeyError vs device_map='auto' PLE device mismatch." + "reason": "Same model family, but different failures in different paths (FSDP training KeyError vs device-map forward mismatch)." }, { "left": "issue:43366", "right": "issue:43531", "accept": false, - "reason": "Not the same underlying bug; one is GGUF architecture support, the other is sliding_window handling in Qwen3-MoE." + "reason": "Unrelated Qwen3-MoE sliding-window bug vs GGUF gpt-oss support request." }, { "left": "issue:44155", "right": "issue:45584", "accept": false, - "reason": "Different subsystems and symptoms: AudioFlamingo3 batch leakage vs Whisper empty-transcription generation failure." + "reason": "Different audio/ASR paths and distinct symptoms; not the same underlying bug." }, { "left": "issue:43950", "right": "issue:45834", "accept": false, - "reason": "One is a from_pretrained buffer-corruption regression, the other is a long-input OCR index error in Kosmos2.5." + "reason": "Non-persistent-buffer regression vs long-OCR index error; unrelated code paths." }, { "left": "issue:44560", "right": "issue:45685", "accept": false, - "reason": "Unrelated failures: Qwen3-vl video StopIteration vs MPS histogram not implemented for int in moe code." + "reason": "Different subsystems and errors; no plausible single fix for both." }, { "left": "issue:43824", "right": "issue:43957", "accept": false, - "reason": "Both are loading-related, but one is an import error for a missing class and the other is meta-device loading breakage." + "reason": "Import surface issue vs meta-device loading bug; different concrete problems." }, { "left": "issue:43550", "right": "issue:44910", "accept": false, - "reason": "Different concrete bugs: torch.compile+SDPA in Bamba vs FlashAttention illegal access from Qwen3.5 position_ids handling." + "reason": "Separate attention-path bugs in different models; not mergeable into one fix." }, { "left": "issue:42915", "right": "issue:43957", "accept": false, - "reason": "Different problems: FineGrainedFP8Config on Qwen3-MoE vs meta-device loading regressions." + "reason": "FP8/Qwen3-MoE failure is unrelated to meta-device loading." }, { "left": "issue:43366", "right": "issue:44589", "accept": false, - "reason": "No shared code-path evidence; GGUF support vs Float8 storage lookup failure." + "reason": "GGUF support request vs float8 storage lookup error; unrelated issues." }, { "left": "issue:43645", "right": "issue:43646", "accept": true, - "reason": "Same underlying regression in custom model initialization; the notebook detail is just a repro environment." + "reason": "Both describe the same Transformers v5 custom-model initialization regression; one is just the notebook-specific report." }, { "left": "issue:44560", "right": "issue:44918", "accept": false, - "reason": "Distinct Qwen3.5 issues: video inference StopIteration vs unpacking embeddings for TRL SFT." + "reason": "Video StopIteration and SFT embedding unpacking are different Qwen3.5 failures." }, { "left": "issue:42617", "right": "issue:43366", "accept": false, - "reason": "Unrelated: 3d_parallel.py runtime failure vs GGUF model-architecture support." + "reason": "No shared bug; one is a script/runtime failure, the other is GGUF architecture support." }, { "left": "issue:36010", "right": "issue:42915", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to Qwen3-MoE FP8 loading." + "reason": "ImportError for GenerationMixin is unrelated to Qwen3-MoE FP8 behavior." }, { "left": "issue:44661", "right": "issue:45823", "accept": false, - "reason": "Different bug classes: tokenizer mapping/new-model registration logic vs Gemma4 PLE device mismatch." + "reason": "Tokenizer-mapping model discovery bug vs Gemma4 PLE device mismatch; different code paths." }, { "left": "issue:44560", "right": "issue:44805", "accept": false, - "reason": "Different concrete failures and model paths; no evidence of one shared root cause." + "reason": "Different error types and model-specific failures." }, { "left": "issue:43632", "right": "issue:45834", "accept": false, - "reason": "Transformers v5 flag/init regressions are unrelated to Kosmos2.5 long-input indexing." + "reason": "_is_hf_initialized regression is unrelated to Kosmos2.5 long-input indexing." }, { "left": "issue:38175", "right": "issue:43994", "accept": true, - "reason": "Both describe SigLIP2-base producing broken inference outputs; these look like the same underlying model/processor regression." + "reason": "Same SigLIP2 model and same wrong-output symptom family; these look like duplicate reports of one inference bug." }, { "left": "issue:43638", "right": "issue:44661", "accept": false, - "reason": "Different issues: deepspeed zero3/BertModel indexing vs add-new-model-like tokenizer-mapping failure." + "reason": "DeepSpeed ZeRO-3 Bert indexing failure is unrelated to tokenizer mapping-name handling." }, { "left": "issue:43596", "right": "issue:44661", "accept": false, - "reason": "One is a zero-size tensor index error during ZeRO-3 init; the other is model-registration logic." + "reason": "Same as above: ZeRO-3/Bert init failure vs tokenizer mapping bug." }, { "left": "issue:44661", "right": "issue:44805", "accept": false, - "reason": "No shared bug: add-new-model-like/tokenizer mapping vs a mask shape mismatch index error." + "reason": "Tokenizer-mapping issue and mask-shape IndexError are not the same bug." }, { "left": "issue:44877", "right": "issue:45030", "accept": false, - "reason": "Both involve config validation, but the affected model families and failures differ and the shared root cause is not proven." + "reason": "Both involve config validation, but the titles point to different models and distinct failing checks." }, { "left": "issue:43317", "right": "issue:44756", "accept": false, - "reason": "Different symptoms and paths: dequantized model offload loading vs mmap OOM on Strix Halo." + "reason": "Device-map/offload loading failure is unrelated to mmap OOM on Strix Halo." }, { "left": "issue:43644", "right": "issue:44534", "accept": true, - "reason": "Exact duplicate regression: both report non-persistent buffers being filled with junk in Transformers v5." + "reason": "Exact same non-persistent-buffer junk regression, just reported with different wording." }, { "left": "issue:38175", "right": "issue:39692", "accept": false, - "reason": "Related SigLIP2 area, but one is bad model outputs while the other is a documentation/example mismatch and quantization failure." + "reason": "Both are SigLIP2-related, but one is bad model outputs and the other is a docs/example mismatch plus quantization failure." }, { "left": "issue:42491", "right": "issue:43720", "accept": false, - "reason": "Different model stacks and failures: Qwen3_MoE LoRA compatibility vs BitNet packed-weight loading." + "reason": "LoRA/qwen3_moe version compatibility and BitNet packed-weight loading are unrelated." }, { "left": "issue:41720", "right": "issue:42915", "accept": false, - "reason": "Both are Qwen, but one is an auto device-map cuda assert and the other is an FP8 configuration failure." + "reason": "Different Qwen3 failures: device-map CUDA assert vs FineGrainedFP8Config." }, { "left": "issue:44960", "right": "issue:44977", "accept": false, - "reason": "Different models and subsystems; no concrete shared code-path." + "reason": "Different GLM/Qwen3.5 issues with different failure modes." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": true, + "reason": "Both are reports that current/latest Transformers cannot load specific BERT-family models; likely the same loader regression." }, { "left": "issue:45313", "right": "issue:45357", "accept": false, - "reason": "Load-time ZeRO-3 language_model weight issue vs save_pretrained visual-encoder key regression." + "reason": "Different Qwen3.5 issues affecting separate components (weight loading vs save_pretrained key handling)." }, { - "left": "issue:44451", - "right": "issue:44488", + "left": "issue:42915", + "right": "issue:44756", "accept": false, - "reason": "Similar symptom class, but different model checkpoints and insufficient evidence that they are the same bug." + "reason": "FP8 Qwen3-MoE failure and Strix Halo mmap OOM are unrelated." }, { - "left": "issue:43644", - "right": "issue:44534", + "left": "issue:44910", + "right": "issue:44977", "accept": true, - "reason": "Exact duplicate regression: both report non-persistent buffers being filled with junk in Transformers v5." + "reason": "Both are Qwen3.5 flash-attention generation failures and likely the same underlying attention/position-id bug." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative overall. The accepted soft matches are limited to clearly related issue pairs by title/symptom, and the rejected pairs generally avoid overclaiming shared root cause. No mergeability concerns here since this is an issue-cluster review, not a PR pair." + "feedback": "The summary is grounded in the packet, and the accepted soft edges are reasonably conservative given the titles and descriptions. The canonical issue choice for the non-persistent-buffer regression is well supported by the near-exact duplicate pair, and the other accepted groups are plausible duplicate/near-duplicate bug reports rather than clearly separate failures." }, "evaluator_used": true, "retried": false, @@ -1849,8 +1924,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -1922,8 +1997,8 @@ "issue:44164|issue:44291", "issue:36010|issue:39404", "issue:45276|issue:45335", - "issue:43404|issue:43454", "issue:43425|issue:43957", + "issue:43404|issue:43454", "issue:41720|issue:44155", "issue:44402|issue:44488", "issue:39692|issue:44960", @@ -1932,8 +2007,8 @@ "issue:43873|issue:44387", "issue:44291|issue:44756", "issue:45663|issue:45859", - "issue:43335|issue:45663", "issue:44155|issue:45823", + "issue:43335|issue:45663", "issue:44292|issue:44912", "issue:44387|issue:44756", "issue:43957|issue:44756", @@ -1941,20 +2016,20 @@ "issue:45276|issue:45292", "issue:44164|issue:44756" ], - "prepared_review_unit_hash": "13b9cc90d1d0348d8a2ae1ab960e13a7adde0a15a3532a90da0b27908c1c8f84", + "prepared_review_unit_hash": "123b51e1b86a1ede2c73727e715132f9102ce9430e3d037b65eced5ebeb6dbc1", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 36, - "serialized_chars": 38648, + "serialized_chars": 38645, "estimated_input_tokens": 9662, "estimated_eval_tokens": 19580 }, @@ -1963,234 +2038,234 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a heterogeneous cluster of unrelated Transformers issues, with two likely duplicate subclusters: resize_token_embeddings regressions and lm_head weight-tying bugs in multimodal models.", - "confidence": 0.74, - "canonical_issue_reason": "issue:45276 is the clearest representative of the resize_token_embeddings regression cluster and has the strongest duplicate links to closely related embedding/output-head reports.", + "summary": "Most items are unrelated single-issue reports; the only clear duplicate family is around `resize_token_embeddings` not propagating resized embeddings/output heads across model variants. I would keep issue 45276 as the representative for that family.", + "confidence": 0.72, + "canonical_issue_reason": "Issue 45276 is the most specific and central report in the resize-token-embeddings family: it names the failing API and covers both input and output embedding propagation, which subsumes the more generic 45292 and the model-specific 45335 symptoms.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45276 best captures the shared underlying embedding-resize bug and is the most central issue among the likely duplicates.", + "best_issue_reason": "Issue 45276 is the best global issue representative because it is detailed enough to anchor the underlying bug while still broad enough to represent the closely related resize-token-embeddings reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43708", "right": "issue:44075", "accept": false, - "reason": "Different bugs: checkpoint resume step calculation vs optimizer argument handling." + "reason": "Different bugs: checkpoint resume step accounting vs SGD optimizer arguments not being applied." }, { "left": "issue:42491", "right": "issue:45276", "accept": false, - "reason": "LoRA checkpoint compatibility on Qwen3 MoE is unrelated to token embedding resize behavior." + "reason": "Both involve model usage/regression, but the concrete failures are unrelated: Qwen3 MoE LoRA compatibility vs token embedding resizing." }, { "left": "issue:43854", "right": "issue:43931", "accept": false, - "reason": "Different model-loading failures on different models with no shared concrete code path." + "reason": "Both mention Qwen3-VL loading, but one is a generic unit-test load failure and the other is a specific shape-mismatch error; not clearly the same bug." }, { "left": "issue:45488", "right": "issue:45812", "accept": false, - "reason": "Both are tokenizer regressions, but one is pre-tokenizer override and the other is wrong token IDs; not the same bug." + "reason": "Same broad tokenizer area, but different model families and different failure modes; too weak to treat as the same underlying bug." }, { "left": "issue:43257", "right": "issue:43854", "accept": false, - "reason": "Qwen3 MoE weight conversion/offload issue vs GLM-4.7 loading failure are unrelated." + "reason": "Different problems: MoE weight conversion under accelerate+deepspeed vs model loading/test issues for Qwen3-VL." }, { "left": "issue:44488", "right": "issue:44991", "accept": false, - "reason": "Both involve tokenizer loading, but the affected models and failure modes differ too much to treat as duplicates." + "reason": "Both are tokenizer-loading regressions, but they affect different models and likely different code paths." }, { "left": "issue:39692", "right": "issue:43994", "accept": false, - "reason": "Same family, but one is a docs example with multiple errors and quantization failure; the other is a runtime nonsense-output report." + "reason": "SigLIP2 doc/example errors and nonsensical runtime outputs are related to the same model family, but the reported bugs are not the same concrete issue." }, { "left": "issue:43644", "right": "issue:43645", "accept": false, - "reason": "Both touch Transformers 5 init/loading, but they describe different symptoms and code paths." + "reason": "Separate 5.0 regressions: junk-filled non-persistent buffers vs Jupyter custom model initialization." }, { "left": "issue:43856", "right": "issue:43873", "accept": false, - "reason": "Training memory use and quantized offloading behavior are not the same underlying issue." + "reason": "Different quantization/offloading/memory complaints; not the same underlying bug." }, { "left": "issue:42491", "right": "issue:43931", "accept": false, - "reason": "Unrelated issues: Qwen3 MoE LoRA reuse vs Qwen3-VL weight-shape mismatch." + "reason": "Different concrete failures: LoRA compatibility across versions vs weight-shape mismatch while loading a VL model." }, { "left": "issue:45362", "right": "issue:45406", "accept": false, - "reason": "Different failures in different model/processor stacks." + "reason": "Both are chat/processor-related model issues, but one is Qwen3.5 chat crashes and the other is a Gemma4 processor AttributeError." }, { "left": "issue:43856", "right": "issue:44387", "accept": false, - "reason": "Both memory-related, but one is MoE training efficiency and the other is int4 reserved-memory/OOM behavior." + "reason": "Both concern memory, but one is Qwen3 MoE training efficiency and the other is int4 quantization reserved-memory/OOM regression." }, { "left": "issue:44960", "right": "issue:46050", "accept": false, - "reason": "Different model families and different save/load regression types." + "reason": "Different model/version topics: GLM5 versus a sinq save/load corruption report." }, { "left": "issue:42491", "right": "issue:45292", "accept": false, - "reason": "Qwen3 MoE LoRA compatibility is unrelated to resize_token_embeddings behavior." + "reason": "Unrelated issues: Qwen3 MoE LoRA compatibility vs embedding resize propagation." }, { "left": "issue:44910", "right": "issue:45588", "accept": false, - "reason": "Both involve flash attention, but the concrete crash causes are different." + "reason": "Both touch flash attention, but the failures are distinct code paths: packed-sequence misinterpretation vs `s_aux=None` AttributeError." }, { "left": "issue:44164", "right": "issue:44291", "accept": false, - "reason": "Save/from_pretrained extra_state handling is unrelated to init_empty_weights argument plumbing." + "reason": "Both involve model loading/saving internals, but extra_state handling and `_is_hf_initialized` argument handling are different bugs." }, { "left": "issue:36010", "right": "issue:39404", "accept": false, - "reason": "ImportError for GenerationMixin vs Whisper pipeline return_language regression are unrelated." + "reason": "ImportError for `GenerationMixin` is unrelated to Whisper `return_language` pipeline behavior." }, { "left": "issue:45276", "right": "issue:45335", "accept": true, - "reason": "Same underlying resize_token_embeddings regression affecting model-specific embeddings/output heads." - }, - { - "left": "issue:43404", - "right": "issue:43454", - "accept": true, - "reason": "Both report missing lm_head weight tying in multimodal generation models, a shared concrete fix path." + "reason": "Both report the same underlying `resize_token_embeddings` propagation bug: resized embeddings are not reflected in model-specific output/input embedding modules." }, { "left": "issue:43425", "right": "issue:43957", "accept": false, - "reason": "Torch version compatibility vs meta-device loading regression are different problems." + "reason": "Torch compatibility and meta-device loading failures are different regressions." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both are weight-tying issues, but for different models and different symptoms; not the same concrete bug." }, { "left": "issue:41720", "right": "issue:44155", "accept": false, - "reason": "Qwen3 device-map assert and AudioFlamingo3 track leakage are unrelated." + "reason": "Qwen3 auto device-map CUDA assert and AudioFlamingo3 batched inference token leakage are unrelated." }, { "left": "issue:44402", "right": "issue:44488", "accept": false, - "reason": "Tokenizer vocab mismatch vs tokenizer loading failure are not the same bug." + "reason": "Tokenizer vocab-size mismatch for one model is not the same as a separate model-loading failure for another." }, { "left": "issue:39692", "right": "issue:44960", "accept": false, - "reason": "SigLIP2 docs/runtime issue is unrelated to GLM5 loading." + "reason": "SigLIP2 documentation/model issues are unrelated to the GLM5 report." }, { "left": "issue:42947", "right": "issue:45704", "accept": false, - "reason": "Gradient checkpointing bug and apex RMSNorm memory leak are unrelated." + "reason": "Gradient checkpointing with PEFT LoRA and an apex RMSNorm memory leak are different bugs." }, { "left": "issue:45663", "right": "issue:45759", "accept": false, - "reason": "Different model architectures and different failure modes." + "reason": "Gemma4 FSDP2 KeyError and Qwen3.5 multimodal config unwrapping are different failure modes." }, { "left": "issue:43873", "right": "issue:44387", "accept": false, - "reason": "Offloading/quantization behavior vs int4 reserved-memory OOM are related only broadly, not the same concrete issue." + "reason": "Quantization/offloading issue vs int4 reserved-memory regression; same area, different bugs." }, { "left": "issue:44291", "right": "issue:44756", "accept": false, - "reason": "Init-empty-weights loading bug vs mmap OOM on Strix Halo are unrelated." + "reason": "Meta-init loading argument mismatch and mmap-related OOM are unrelated." }, { "left": "issue:45663", "right": "issue:45859", "accept": false, - "reason": "Gemma4 FSDP2 keyerror and Qwen3 MoE CPU-offload incompatibility are different bugs." + "reason": "Gemma4 FSDP2 attention state handling and Qwen3_5Moe CPU-offload incompatibility are distinct." }, { - "left": "issue:43335", - "right": "issue:45663", + "left": "issue:44155", + "right": "issue:45823", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug is unrelated to Gemma4 attention state handling." + "reason": "AudioFlamingo3 batched inference leakage and Gemma4 PLE device mismatch are unrelated." }, { - "left": "issue:44155", - "right": "issue:45823", + "left": "issue:43335", + "right": "issue:45663", "accept": false, - "reason": "AudioFlamingo token leak and Gemma4 device mismatch are unrelated." + "reason": "Sparse-layer creation in SwitchTransformers and Gemma4 FSDP2 shared_kv_state rebuilding are unrelated." }, { "left": "issue:44292", "right": "issue:44912", "accept": false, - "reason": "Different quantization/loading regressions on different models and formats." + "reason": "Different quantization/model-loading problems for different models." }, { "left": "issue:44387", "right": "issue:44756", "accept": false, - "reason": "Both are memory issues, but one is quantization-driven OOM and the other is mmap-related." + "reason": "Different causes of OOM: int4 reserved memory growth vs mmap on Strix Halo." }, { "left": "issue:43957", "right": "issue:44756", "accept": false, - "reason": "Meta-device model loading regression vs mmap disablement for OOM are not the same bug." + "reason": "Meta-device loading breakage and mmap/OOM are different issues." }, { "left": "issue:43854", "right": "issue:45216", "accept": false, - "reason": "Model load failure vs save_pretrained checkpoint regression are distinct." + "reason": "Unit-test model loading failure for GLM-4.7-Flash vs Qwen3.5 save_pretrained regression are different bugs." }, { "left": "issue:45276", "right": "issue:45292", "accept": true, - "reason": "Same resize_token_embeddings/output_embeddings regression described from two angles." + "reason": "Same underlying `resize_token_embeddings` bug: resized embeddings are not reflected in output embeddings." }, { "left": "issue:44164", "right": "issue:44756", "accept": false, - "reason": "extra_state serialization bug and Strix Halo mmap OOM are unrelated." + "reason": "Extra_state serialization and Strix Halo mmap OOM are unrelated." } ] }, "evaluator_result": { "accept": true, - "feedback": "The clustering is grounded in the packet and the summary stays conservative. The two accepted duplicate subclusters are well supported by the issue titles and high soft-similarity pairs, and the rejected pairs are generally phrased cautiously without overclaiming evidence." + "feedback": "The summary is grounded in the packet: 45276, 45292, and 45335 clearly form the resize_token_embeddings family, and choosing 45276 as the representative is conservative and well supported by the titles. The soft-edge verdicts are also generally cautious, mostly rejecting pairs with different model families or failure modes rather than overclaiming duplicates. I don\u2019t see an overstatement that would justify rejection." }, "evaluator_used": true, "retried": false, @@ -2206,19 +2281,17 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ - "issue:36032", "issue:39404", "issue:42947", "issue:43257", "issue:43299", "issue:43381", "issue:43425", - "issue:43452", "issue:43596", "issue:43611", "issue:43645", @@ -2258,6 +2331,7 @@ "issue:45684", "issue:45698", "issue:45701", + "issue:45759", "issue:45823" ], "soft_pairs": [ @@ -2291,228 +2365,222 @@ "issue:44589|issue:45084", "issue:43596|issue:44683", "issue:39404|issue:44671", - "issue:36032|issue:43452" + "issue:44164|issue:45759" ], - "prepared_review_unit_hash": "0f1b0e6e4837478b67b9e26a3bcd988d68d52304a7fdc5b48ae8d00c09ec2813", + "prepared_review_unit_hash": "377f2324901e6c78faabdbeffda8ff3d57607ee138bc54e61d0ab6f549d2a618", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 48, - "item_count": 48, + "node_count": 47, + "item_count": 47, "soft_pair_count": 31, - "serialized_chars": 37067, - "estimated_input_tokens": 9267, - "estimated_eval_tokens": 18790 + "serialized_chars": 36466, + "estimated_input_tokens": 9117, + "estimated_eval_tokens": 18490 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most candidate pairs are only loosely related and describe different bugs. The only strong duplicate-looking pair is the `resize_token_embeddings` regression, where the generic report is the best canonical issue.", + "summary": "This cluster is mostly a grab-bag of unrelated Transformers regressions; only the resize_token_embeddings pair looks like a true duplicate. Most other pairs share a subsystem or vague symptom but not the same underlying bug.", "confidence": 0.88, - "canonical_issue_reason": "issue:45292 is the broader, model-agnostic report of the same resize_token_embeddings propagation bug; issue:45335 is a specific t5gemma manifestation.", + "canonical_issue_reason": "issue:45292 is the broadest and most central formulation of the resize_token_embeddings bug; issue:45335 reads like a model-specific manifestation of the same problem.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45292 best represents the underlying defect because it states the generic failure mode and is not limited to one architecture.", + "best_issue_reason": "issue:45292 is the best canonical issue because it describes the generic behavior regression and subsumes the more specific t5gemma report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43381", "right": "issue:44928", "accept": false, - "reason": "Different failures: eval-mode gradient checkpointing vs RLHF NaN from 3D position_ids / SDPA fallback." + "reason": "Both involve gradient-related training behavior, but one is an eval-mode checkpointing restriction and the other is a Qwen3.5 RLHF overflow/SDPA fallback bug." }, { "left": "issue:45684", "right": "issue:45698", "accept": false, - "reason": "Both involve save/load, but one is a file-permissions problem and the other is wrong custom-module resolution." + "reason": "Both mention save_pretrained/from_pretrained, but one is about read-only file permissions during save and the other is about loading the wrong custom module." }, { "left": "issue:42947", "right": "issue:43381", "accept": false, - "reason": "Related to gradient checkpointing, but one is ineffective LoRA checkpointing and the other is eval-mode incompatibility." + "reason": "LoRA gradient checkpointing ineffective is a different bug from gradient checkpointing being disallowed in eval mode." }, { "left": "issue:44479", "right": "issue:44877", "accept": false, - "reason": "Different subsystems: video-input regression in Qwen VL models vs strict-config loading failure for granite_speech." + "reason": "Different model families and failure modes: video-input regression vs strict config blocking granite_speech loading." }, { "left": "issue:45588", "right": "issue:45663", "accept": false, - "reason": "Both are attention/training bugs, but the concrete failures and code paths are unrelated." + "reason": "A flash_attention crash on sink-less models is not the same as a Gemma4 FSDP2 shared_kv_states KeyError." }, { "left": "issue:43646", "right": "issue:43708", "accept": false, - "reason": "Custom model init in notebooks is unrelated to Trainer checkpoint-resume max_steps calculation." + "reason": "Custom model initialization in notebooks is unrelated to resume_from_checkpoint max_steps calculation." }, { "left": "issue:43645", "right": "issue:44164", "accept": false, - "reason": "Different underlying bugs: notebook-defined custom models vs save/from_pretrained extra_state handling." + "reason": "Notebook-defined custom model initialization and extra_state save/load handling are different code paths and bugs." }, { "left": "issue:43828", "right": "issue:43957", "accept": false, - "reason": "Both are loading/runtime mismatches, but autocast dtype mismatch and meta-device loading are different bugs." + "reason": "Both are dtype/device-related, but one is an autocast mismatch for Phi-tiny-MoE and the other is a meta-device loading regression." }, { "left": "issue:45292", "right": "issue:45335", "accept": true, - "reason": "Same underlying resize_token_embeddings regression; 45335 is the t5gemma-specific instance of the generic output-embedding issue." + "reason": "These describe the same resize_token_embeddings regression; 45335 is a model-specific instance of the broader output_embeddings issue." }, { "left": "issue:44402", "right": "issue:44451", "accept": false, - "reason": "Tokenizer vocab-size mismatch is not the same as an inability to load ScandiBERT in the latest version." + "reason": "Both concern loading/tokenizer oddities, but the affected models and likely root causes are different." }, { "left": "issue:43425", "right": "issue:43828", "accept": false, - "reason": "Torch-version incompatibility is too broad and does not match the autocast dtype mismatch bug." + "reason": "A torch version compatibility issue is not the same as a model-specific dtype mismatch under autocast." }, { "left": "issue:43257", "right": "issue:44863", "accept": false, - "reason": "Both are model-loading issues, but they concern different models and different failure mechanisms." + "reason": "Both are load failures for large models, but they involve different architectures and different failure mechanisms." }, { "left": "issue:44589", "right": "issue:44756", "accept": false, - "reason": "Float8 storage lookup failure is unrelated to mmap-induced OOM on Strix Halo." + "reason": "Missing Float8 storage and disabling mmap for OOM are unrelated problems." }, { "left": "issue:44683", "right": "issue:45823", "accept": false, - "reason": "Compiled flex_attention on newer torch versions is unrelated to Gemma4 device-map forward mismatch." + "reason": "Torch flex_attention compilation on >=2.9 and Gemma4 PLE device-map mismatch are different bugs." }, { "left": "issue:39404", "right": "issue:45701", "accept": false, - "reason": "Whisper pipeline return_language regression is not the same as a tokenizer-version behavioral change." + "reason": "Whisper return_language pipeline regression is unrelated to tokenization behavior changes." }, { "left": "issue:43425", "right": "issue:45823", "accept": false, - "reason": "General torch incompatibility does not describe the Gemma4 device mismatch bug." + "reason": "General torch incompatibility and Gemma4 device mismatch do not look like the same bug." }, { "left": "issue:43425", "right": "issue:43716", "accept": false, - "reason": "The issues are about different problems: torch compatibility vs image preprocessor/model dtype mismatch." + "reason": "Torch incompatibility and Mistral-3 image/text dtype mismatch are different issues." }, { "left": "issue:43611", "right": "issue:43645", "accept": false, - "reason": "Both touch custom-model loading in v5, but one is base_model_prefix handling and the other is notebook-defined class initialization." + "reason": "Both involve custom models, but one is a base_model_prefix loading regression and the other is Jupyter notebook initialization behavior." }, { "left": "issue:43906", "right": "issue:44530", "accept": false, - "reason": "An isolated reproduction of another issue is not the same as the PagedAttentionCache linear_attention crash." + "reason": "An isolated reproduction of another issue and a PagedAttentionCache linear_attention crash are not the same concrete bug." }, { "left": "issue:44530", "right": "issue:45084", "accept": false, - "reason": "Different failure modes: cache group-type handling vs compilation of non-template nodes." + "reason": "PagedAttentionCache invalid group type and compiler non-template-node errors are unrelated failures." }, { "left": "issue:39404", - "right": "issue:44912", + "right": "issue:44671", "accept": false, - "reason": "Whisper pipeline return_language regression is unrelated to MXFP4 quantization fallback." + "reason": "Whisper pipeline language return and CamemBERT masked LM prediction drift are unrelated model-specific regressions." }, { "left": "issue:43901", "right": "issue:44509", "accept": false, - "reason": "Both are docs-related, but they concern different documentation mismatches and different removed/changed behaviors." + "reason": "Documentation mismatch about return_all_scores is unrelated to docs mentioning removed pipeline tasks." }, { "left": "issue:44265", "right": "issue:44898", "accept": false, - "reason": "Different model/export failures: torch.export with torch_compilable_check vs Perceiver image resizing." + "reason": "torch.export failing on torch_compilable_check is a different issue from Perceiver image classification failing with interpolate_pos_encoding." }, { "left": "issue:44479", "right": "issue:44805", "accept": false, - "reason": "Video-input regression in Qwen VL models is unrelated to the mask/tensor shape IndexError." + "reason": "A video-input regression for Qwen/VL models is not the same as a generic mask-shape IndexError." }, { "left": "issue:43994", "right": "issue:44079", "accept": false, - "reason": "Incorrect SigLIP2 results are not clearly the same bug as ModelOutput key assignment when a key was previously None." + "reason": "Nonsensical predictions from a specific model are not the same as ModelOutput key assignment logic breaking after None values." }, { "left": "issue:43425", "right": "issue:44918", "accept": false, - "reason": "Torch 2.10 incompatibility is unrelated to the TRL SFT embedding unpacking failure." + "reason": "Torch compatibility and Qwen3.5 embedding unpacking are different regressions." }, { "left": "issue:43299", "right": "issue:43931", "accept": false, - "reason": "Both involve Qwen3-VL loading, but one is a version regression for MoE models and the other is a shape mismatch on a specific checkpoint." + "reason": "Both are Qwen3-VL loading problems, but the titles indicate different failure modes and not enough evidence of the same root cause." }, { "left": "issue:44589", "right": "issue:45084", "accept": false, - "reason": "Float8 storage resolution and non-template-node compilation are separate bugs." + "reason": "A storage-object lookup failure and a compiler template-node error are unrelated." }, { "left": "issue:43596", "right": "issue:44683", "accept": false, - "reason": "DeepSpeed ZeRO-3/BertModel indexing error is unrelated to compiled flex_attention on torch >= 2.9." - }, - { - "left": "issue:43906", - "right": "issue:44530", - "accept": false, - "reason": "These describe different problems; the former is a reproduction report, the latter is a Qwen3.5 cache crash." + "reason": "Zero3 init index error with BertModel is unrelated to flex_attention compilation on torch >=2.9." }, { - "left": "issue:36032", - "right": "issue:43452", + "left": "issue:44164", + "right": "issue:45759", "accept": false, - "reason": "Tokenizer add_special_tokens conflict is unrelated to gguf_file breaking tokenizer/model loading." + "reason": "extra_state save/load handling and unwrapping text_config for composite Qwen configs are different loading/saving bugs." } ] }, "evaluator_result": { "accept": true, - "feedback": "The analyst output is conservative and grounded in the packet. The selected canonical pair, issue:45292 vs issue:45335, is the clearly strongest duplicate-looking match from the titles, and the remaining soft-edge rejections are appropriately cautious given the available evidence. No overstatement stands out." + "feedback": "The summary is grounded in the packet and stays conservative. The resize_token_embeddings pair is the only clearly duplicate-looking match, and the other soft-edge verdicts appropriately avoid overclaiming despite superficial subsystem overlap." }, "evaluator_used": true, "retried": false, @@ -2528,8 +2596,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -2540,7 +2608,9 @@ "issue:40444", "issue:40990", "issue:42915", + "issue:43262", "issue:43317", + "issue:43377", "issue:43421", "issue:43452", "issue:43482", @@ -2577,12 +2647,11 @@ "issue:45325", "issue:45507", "issue:45584", - "issue:45759", "issue:45859", "issue:45987" ], "soft_pairs": [ - "issue:44164|issue:45759", + "issue:36032|issue:43452", "issue:44671|issue:44991", "issue:32090|issue:39404", "issue:43856|issue:44756", @@ -2613,238 +2682,242 @@ "issue:36032|issue:43645", "issue:44987|issue:44991", "issue:45084|issue:45507", - "issue:43883|issue:44991" + "issue:43883|issue:44991", + "issue:43262|issue:43377" ], - "prepared_review_unit_hash": "3f5056595bb8750275b0e9274a03f069bb6d2f4aab689b4654f3315d5df100ea", + "prepared_review_unit_hash": "e12f91e6f5805f48aa745d7c27fbb87d4d47f2c0731dfff93e07b6f7c19ea290", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 32, - "serialized_chars": 36622, - "estimated_input_tokens": 9156, - "estimated_eval_tokens": 18568 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37542, + "estimated_input_tokens": 9386, + "estimated_eval_tokens": 19028 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly conservative rejections; the only pair that needed adjustment is the GGUF report, which is too weakly evidenced to treat as a confirmed duplicate/root-cause match.", - "confidence": 0.57, - "canonical_issue_reason": "issue:43452 is the broader GGUF-loading report, but the evidence here does not establish that it fully subsumes the Qwen2.5-specific failure in issue:43482.", + "summary": "This cluster is mostly a loose set of Transformers v5 regressions and model/tokenizer-loading bugs, but the items are not the same underlying defect. I would keep one broad v5 regression report as the representative issue, and reject all soft duplicate links.", + "confidence": 0.41, + "canonical_issue_reason": "issue:43646 is the broadest and most representative title in the cluster: a generic Transformers v5 custom-model initialization regression. It best matches the cluster\u2019s dominant theme, even though the overall set is still heterogeneous.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43452 is still the best umbrella candidate among the two GGUF issues, but only as a tentative representative because the shared root cause is not confirmed from the available text.", + "best_issue_reason": "issue:43646 is the best single issue to keep because it is broad, central to the v5 regression theme, and not overly tied to one model or one narrow symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:44164", - "right": "issue:45759", + "left": "issue:36032", + "right": "issue:43452", "accept": false, - "reason": "Different API paths and failure modes: extra_state save/load handling vs text_config unwrapping for composite multimodal configs." + "reason": "Different failure modes: T5 tokenizer attribute conflict vs gguf_file loading path breakage." }, { "left": "issue:44671", "right": "issue:44991", "accept": false, - "reason": "Both are v5 loading regressions, but they hit different models and different failure modes (masked-LM outputs vs tokenizer loading)." + "reason": "Both are model-loading regressions, but one is masked LM prediction quality and the other is tokenizer loading; not the same bug." }, { "left": "issue:32090", "right": "issue:39404", "accept": false, - "reason": "Unrelated runtime errors in Trainer broadcast vs Whisper pipeline return_language behavior." + "reason": "Trainer GPU broadcast TypeError and Whisper return_language pipeline regression are unrelated." }, { "left": "issue:43856", "right": "issue:44756", "accept": false, - "reason": "Both mention memory/OOM, but one is Qwen3 MoE training efficiency and the other is mmap behavior on Strix Halo; not the same bug." + "reason": "Different symptoms and subsystems: Qwen3 MoE training memory usage vs mmap OOM on Strix Halo." }, { "left": "issue:44991", "right": "issue:45081", "accept": false, - "reason": "Different tokenizer/model loading regressions with different affected models and causes; too broad to merge." + "reason": "Both affect loading, but one is tokenizer loading for a specific model and the other is a Mistral regex patch crash; distinct code paths." }, { "left": "issue:44737", "right": "issue:45859", "accept": false, - "reason": "Both are device/offload issues, but XLNet CPU arange placement and Qwen3.5 MoE rotary embedding offload are distinct code paths." + "reason": "CPU device-placement bug in XLNet vs Qwen3_5Moe rotary embedding offload compatibility; not the same issue." }, { "left": "issue:44164", "right": "issue:44479", "accept": false, - "reason": "Different functionality entirely: extra_state serialization vs video-input regression in multimodal Qwen models." + "reason": "save/from_pretrained extra_state handling is unrelated to the Qwen VL video-input regression." }, { "left": "issue:43582", "right": "issue:43632", "accept": false, - "reason": "AppleSilicon allocator TypeError and _is_hf_initialized flag regression are separate initialization/runtime problems." + "reason": "AppleSilicon allocator warmup TypeError and _is_hf_initialized flag regression are separate internal issues." }, { "left": "issue:44661", "right": "issue:44683", "accept": false, - "reason": "Model-registration mapping bug vs compiled flex_attention torch-version failure; no shared fix path." + "reason": "Tokenizer mapping add-new-model-like failure is unrelated to compiled flex_attention on torch >= 2.9." }, { "left": "issue:44488", "right": "issue:45081", "accept": false, - "reason": "Both are loading failures, but for different models and different mechanisms (model download/loading vs Mistral regex patching)." + "reason": "Different model families and failure points: generic BERT loading vs Mistral tokenizer regex patch." }, { "left": "issue:44488", "right": "issue:44987", "accept": false, - "reason": "Different affected models and symptoms; one is a generic load failure, the other is a transformers>=5.1.0 regression for a specific repo." + "reason": "Both are load failures, but one is a specific model load issue and the other is a different transformers>=5.1 regression." }, { "left": "issue:43611", "right": "issue:43644", "accept": false, - "reason": "Both are v5-related regressions, but base_model_prefix loading and non-persistent buffer initialization are different underlying bugs." + "reason": "Base_model_prefix loading breakage and non-persistent buffer junk are related to v5 internals but are not the same concrete bug." }, { "left": "issue:40990", "right": "issue:43421", "accept": false, - "reason": "Perplexity/training behavior on gpt-oss-20b is unrelated to runtime special-token post-processing updates." + "reason": "Perplexity regression on GPT-OSS is not the same as runtime post-processor update behavior in TokenizersBackend." }, { "left": "issue:43716", "right": "issue:45084", "accept": false, - "reason": "Different domains: multimodal dtype mismatch vs template-node compilation error." + "reason": "Image preprocessor/model dtype mismatch is unrelated to the non-template-node compilation error." }, { "left": "issue:44206", "right": "issue:44220", "accept": false, - "reason": "Both are audio-feature related, but the titles point to different failing entry points; not enough evidence for one bug." + "reason": "These are both audio-feature related, but one is an unsupported center argument regression and the other is a different fbank extraction problem." }, { "left": "issue:36032", "right": "issue:42915", "accept": false, - "reason": "Tokenizer naming conflict and FineGrainedFP8Config failure are unrelated." + "reason": "Tokenizer load error for T5 is unrelated to Qwen3Moe FineGrainedFP8Config failure." }, { "left": "issue:43643", "right": "issue:44530", "accept": false, - "reason": "trust_remote_code config field loss is not the same as PagedAttentionCache failing on linear_attention group type." + "reason": "Missing fields from AutoConfig with trust_remote_code=True is not the same bug as PagedAttentionCache rejecting linear_attention." }, { "left": "issue:34567", "right": "issue:40444", "accept": false, - "reason": "TrainerState token-count tracking and Qwen2.5-VL iterable-dataset finetuning are different issues." + "reason": "TrainerState token counting bug and multi-image IterableDataset finetuning failure are unrelated." }, { "left": "issue:43883", "right": "issue:44488", "accept": false, - "reason": "Missing all_tied_weights_keys in Molmo and tokenizer-loading failure for sleng-bert are unrelated." + "reason": "Missing all_tied_weights_keys on Molmo is not the same as loading cjvt/sleng-bert." }, { "left": "issue:44811", "right": "issue:45584", "accept": false, - "reason": "Both are Whisper-related, but batch_decode skip_special_tokens handling and empty transcription after align_special_tokens are not clearly the same bug." + "reason": "Both are Whisper-related, but one ignores skip_special_tokens and the other fails on empty transcription after align_special_tokens; different code paths." }, { "left": "issue:44451", "right": "issue:44987", "accept": false, - "reason": "Two different model-loading regressions with no evidence of a shared root cause." + "reason": "Different affected assets and symptoms: one cannot load a specific model, the other is a broader transformers>=5.1 load regression." }, { "left": "issue:43541", "right": "issue:45084", "accept": false, - "reason": "Torch dynamo grouped_mm tracing error and template-node compilation error are unrelated." + "reason": "torch dynamo grouped_mm tracing failure is unrelated to template compilation errors." }, { "left": "issue:42915", "right": "issue:44530", "accept": false, - "reason": "Qwen3Moe FineGrainedFP8Config failure and Qwen3.5 PagedAttentionCache invalid group type are distinct problems." + "reason": "Qwen3Moe FP8 config failure and Qwen3.5 PagedAttentionCache linear_attention crash are different bugs." }, { "left": "issue:44464", "right": "issue:45987", "accept": false, - "reason": "Chunked generation inconsistency with compiled forward is not the same as StaticCache.get_seq_length contract mismatch." + "reason": "Chunked generation inconsistency with compiled forward is not the same as StaticCache.get_seq_length() returning the wrong type." }, { "left": "issue:43452", "right": "issue:43482", "accept": false, - "reason": "The GGUF pair is not strong enough here: 43482 is only a Qwen2.5-GGUF loading failure, and the body evidence does not confirm the same root cause as the generic gguf_file breakage in 43452." + "reason": "Both involve loading, but one is gguf_file handling in general and the other is a Qwen2.5-GGUF v5 compatibility regression." }, { "left": "issue:43644", "right": "issue:43646", "accept": false, - "reason": "Both are v5 initialization regressions, but one is about junk-filled non-persistent buffers and the other about custom model initialization broadly; not clearly the same fix." + "reason": "Related v5 breakages, but one is non-persistent buffer initialization and the other is broader custom model initialization; not identical." }, { "left": "issue:43317", "right": "issue:43482", "accept": false, - "reason": "device_map/offload dequantized-model loading is unrelated to GGUF loading failure." + "reason": "gpu+cpu offload dequantized-model loading issue is different from Qwen2.5-GGUF v5 loading failure." }, { "left": "issue:44849", "right": "issue:45325", "accept": false, - "reason": "Hidden-state output bug in Qwen3.5 vs Qwen2.5-VL rope index scaling are different model-specific issues." + "reason": "Qwen3.5 hidden-states output bug and Qwen2.5-VL rope-index scaling bug affect different model behaviors." }, { "left": "issue:36032", "right": "issue:43645", "accept": false, - "reason": "Tokenizer special-token conflict and notebook custom-model initialization regression are unrelated." + "reason": "T5 tokenizer conflict and Jupyter custom-model initialization regression are different problems." }, { "left": "issue:44987", "right": "issue:44991", "accept": false, - "reason": "Both mention loading failures on transformers v5+, but for different repositories and different failure modes." + "reason": "One is a model loading regression, the other is tokenizer loading for a specific checkpoint; same theme but not same bug." }, { "left": "issue:45084", "right": "issue:45507", "accept": false, - "reason": "Template compilation error and GraniteMoEHybrid invalid-method call are different code paths." + "reason": "Template-node compilation failure is unrelated to GraniteMoEHybrid calling an invalid method." }, { "left": "issue:43883", "right": "issue:44991", "accept": false, - "reason": "Molmo tied-weights attribute error and tokenizer loading failure for est-roberta do not share a concrete bug." + "reason": "Molmo missing attribute and est-roberta tokenizer load failure are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug and MIMI padding-mask batch inconsistency are distinct audio issues." } ] }, - "evaluator_result": { - "accept": false, - "feedback": "The GGUF pair is overstated. In the packet, issue:43452 is the generic gguf_file breakage report, but issue:43482 only says Qwen2.5-GGUF loading failed with transformers v5; the body text needed to confirm the same root cause is not present here. That makes the 'root duplicate target' and 'broadest, most general report' claims too strong for this evidence. The other soft-edge rejections are conservative enough, but this accepted pair should be marked uncertain/rejected." - }, - "evaluator_used": true, - "retried": true, - "accepted_nontrivial_soft_edge": false, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -2856,8 +2929,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -2868,7 +2941,6 @@ "issue:42175", "issue:42915", "issue:43089", - "issue:43262", "issue:43296", "issue:43322", "issue:43377", @@ -2911,18 +2983,17 @@ "issue:45987" ], "soft_pairs": [ - "issue:43262|issue:43377", "issue:43089|issue:45987", "issue:43404|issue:44466", "issue:43646|issue:44534", "issue:43645|issue:44534", - "issue:43322|issue:43540", "issue:44403|issue:45081", + "issue:43322|issue:43540", "issue:43994|issue:44610", "issue:42915|issue:45084", "issue:43377|issue:43819", - "issue:44451|issue:45081", "issue:44220|issue:44683", + "issue:44451|issue:45081", "issue:44987|issue:45081", "issue:44265|issue:44610", "issue:45216|issue:45357", @@ -2931,8 +3002,8 @@ "issue:44062|issue:44987", "issue:36032|issue:43742", "issue:42915|issue:44589", - "issue:43404|issue:45127", "issue:44479|issue:45325", + "issue:43404|issue:45127", "issue:43296|issue:43531", "issue:40444|issue:45325", "issue:36032|issue:41720", @@ -2940,243 +3011,237 @@ "issue:43425|issue:44403", "issue:43611|issue:44534", "issue:43632|issue:44534", - "issue:42175|issue:45542", "issue:43792|issue:44589", + "issue:42175|issue:45542", "issue:40990|issue:44811", "issue:44683|issue:45020" ], - "prepared_review_unit_hash": "16358e5c9e8bb084d9fdb06821e2e4836684c7f05e1c8f80a6d7f427e4712ec6", + "prepared_review_unit_hash": "4bc30f0f77c319c7e38aa4f028c5082124840cd3ffe0d37e5bcd0f6cea8a8a19", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 37573, - "estimated_input_tokens": 9394, - "estimated_eval_tokens": 19044 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36641, + "estimated_input_tokens": 9161, + "estimated_eval_tokens": 18578 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly noisy and heterogeneous: a handful of v5/regression reports, multimodal/audio/model-loading bugs, and unrelated performance/runtime issues. I only see one clear duplicate-like pair: the two Qwen3.5 save_pretrained regressions.", - "confidence": 0.74, - "canonical_issue_reason": "issue:45216 is the broader Qwen3.5 save_pretrained regression report, while issue:45357 looks like a narrower follow-up about the same incorrect visual-encoder key serialization.", + "summary": "Mostly a mixed set of distinct bugs, with one dominant theme around Transformers v5 model-loading/custom-init regressions plus two smaller duplicate-like subclusters (Qwen3.5 checkpoint saving and Qwen VL rope/position-id handling).", + "confidence": 0.67, + "canonical_issue_reason": "Issue 43646 is the broadest and most central representative of the v5 custom-model initialization/loading regression theme; it is more umbrella-like than the narrower notebook/base_model_prefix/_is_hf_initialized variants.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45216 is the best representative of the only clear duplicate subtopic in the cluster: a Qwen3.5 save_pretrained regression affecting checkpoint correctness.", + "best_issue_reason": "43646 is the best single issue to anchor the main cluster because it captures the general v5 custom-model initialization breakage without being overly model-specific.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43262", - "right": "issue:43377", - "accept": false, - "reason": "Both are audio-related, but one is chat-template sampling rate and the other is batched encoder padding-mask behavior; different bugs and code paths." - }, { "left": "issue:43089", "right": "issue:45987", "accept": false, - "reason": "Generation sync overhead and StaticCache.get_seq_length contract mismatch are unrelated runtime issues." + "reason": "Different problems: generation throughput/sync overhead vs a Cache API return-type contract bug." }, { "left": "issue:43404", "right": "issue:44466", "accept": false, - "reason": "Both mention lm_head/tied weights, but one is a model-loading tie issue and the other is device-dependent save_pretrained serialization; not the same bug." + "reason": "Both involve tied weights/lm_head, but one is a Mistral3 model-specific tying bug and the other is a save_pretrained serialization inconsistency; not the same concrete failure." }, { "left": "issue:43646", "right": "issue:44534", "accept": false, - "reason": "Custom model initialization problems and non-persistent buffer junk are different Transformer v5 regressions." + "reason": "Custom model initialization failure vs non-persistent buffer corruption; same release family, but different code paths and symptoms." }, { "left": "issue:43645", "right": "issue:44534", "accept": false, - "reason": "Notebook/custom-model initialization is unrelated to non-persistent buffer initialization." + "reason": "Jupyter/custom model init regression is unrelated to non-persistent buffer junk filling." }, { - "left": "issue:43322", - "right": "issue:43540", + "left": "issue:44403", + "right": "issue:45081", "accept": false, - "reason": "Different multimodal failures in different models; too broad to be the same underlying bug." + "reason": "Generic loading noise issue and a Mistral tokenizer patch crash are different bugs." }, { - "left": "issue:44403", - "right": "issue:45081", + "left": "issue:43322", + "right": "issue:43540", "accept": false, - "reason": "Noise while loading a transformer is not the same as a tokenizer regex patch crash." + "reason": "Different multimodal models and different load/process failures." }, { "left": "issue:43994", "right": "issue:44610", "accept": false, - "reason": "Both are vision-model issues, but one is bad outputs from AutoModel/pipeline and the other is a processor size mismatch; not the same code path." + "reason": "Different vision-model issues: nonsensical AutoModel/pipeline outputs vs processor size mismatch." }, { "left": "issue:42915", "right": "issue:45084", "accept": false, - "reason": "FineGrainedFP8Config failure and template compilation error are unrelated." + "reason": "Qwen3Moe FP8 configuration failure is unrelated to the template-node compilation error." }, { "left": "issue:43377", "right": "issue:43819", "accept": false, - "reason": "Both are audio-model bugs, but batched-vs-single padding mask behavior is different from DAC.from_latents vs forward mismatch." + "reason": "Both are audio-model correctness bugs, but one is padding-mask batching and the other is STE/latent decoding; not the same bug." }, { - "left": "issue:44451", - "right": "issue:45081", + "left": "issue:44220", + "right": "issue:44683", "accept": false, - "reason": "A model-loading failure for ScandiBERT and a Mistral tokenizer regex crash are unrelated." + "reason": "Audio feature extraction bug vs compiled flex_attention incompatibility on newer torch." }, { - "left": "issue:44220", - "right": "issue:44683", + "left": "issue:44451", + "right": "issue:45081", "accept": false, - "reason": "fbank feature extraction and compiled flex_attention failures are different subsystems." + "reason": "Model load failure for ScandiBERT vs tokenizer regex patch crash for Mistral; unrelated." }, { "left": "issue:44987", "right": "issue:45081", "accept": false, - "reason": "Loading a specific remote-code model and a Mistral tokenizer patch crash are different regressions." + "reason": "Different loading regressions affecting different model families and code paths." }, { "left": "issue:44265", "right": "issue:44610", "accept": false, - "reason": "torch.export export failure and OmDet processor size mismatch are not the same bug." + "reason": "torch.export/exportability bug vs OmDet-Turbo processor/image-size mismatch." }, { "left": "issue:45216", "right": "issue:45357", "accept": true, - "reason": "Both report Qwen3.5 save_pretrained producing incorrect checkpoints; 45357 is a narrower description of the same regression." + "reason": "Same model family and same `save_pretrained` serialization regression for Qwen3.5; 45357 is a narrower symptom of the checkpoint corruption reported in 45216." }, { "left": "issue:42915", "right": "issue:44291", "accept": false, - "reason": "FineGrainedFP8Config failure and init_empty_weights/_is_hf_initialized TypeError are distinct issues." + "reason": "Both are init/loading-related, but Qwen3Moe FP8 config and `_is_hf_initialized` argument handling are distinct bugs." }, { "left": "issue:43632", "right": "issue:44756", "accept": false, - "reason": "_is_hf_initialized flag breakage and mmap/OOM handling on Strix Halo are unrelated." + "reason": "Different failures: `_is_hf_initialized` initialization behavior vs mmap/OOM policy on Strix Halo." }, { "left": "issue:44062", "right": "issue:44987", "accept": false, - "reason": "AddedToken special-argument conflict and a remote-code loading regression are different tokenizer/model-loading bugs." + "reason": "Tokenizer AddedToken argument duplication vs a model load failure for `physical-intelligence/fast`." }, { "left": "issue:36032", "right": "issue:43742", "accept": false, - "reason": "Tokenizer add_special_tokens conflict and MobileLLM loading key error are unrelated model-loading failures." + "reason": "T5 tokenizer special-token conflict and MobileLLM key error are unrelated loading bugs." }, { "left": "issue:42915", "right": "issue:44589", "accept": false, - "reason": "FP8 config failure and missing Float8 storage type are different torch/backend compatibility issues." - }, - { - "left": "issue:43404", - "right": "issue:45127", - "accept": false, - "reason": "lm_head tying/serialization and LoRA merge collapse with extended vocab are different weight-handling bugs." + "reason": "Qwen3Moe FP8 config failure is unrelated to the Float8 storage lookup error." }, { "left": "issue:44479", "right": "issue:45325", + "accept": true, + "reason": "Same Qwen VL rope/temporal position-id regression in v5.3.0; 45325 is the specific still-image manifestation of the broader video-input failure in 44479." + }, + { + "left": "issue:43404", + "right": "issue:45127", "accept": false, - "reason": "Both are Qwen vision/video regressions, but one is a general video-input regression and the other is a rope-index temporal scaling bug." + "reason": "Both touch tied embeddings/weights, but one is Mistral3 lm_head tying and the other is LoRA merge with extended vocabulary; not the same bug." }, { "left": "issue:43296", "right": "issue:43531", "accept": false, - "reason": "PaddleOCR-VL loading and Qwen3-MoE sliding_window behavior are unrelated." + "reason": "PaddleOCR-VL load failure and Qwen3-MoE sliding_window behavior are different issues." }, { "left": "issue:40444", "right": "issue:45325", "accept": false, - "reason": "IterableDataset multi-image finetuning and Qwen2.5-VL rope-index scaling are different video/image pipeline bugs." + "reason": "IterableDataset multi-image finetuning failure is unrelated to Qwen2.5-VL temporal position-id scaling." }, { "left": "issue:36032", "right": "issue:41720", "accept": false, - "reason": "T5 tokenizer loading and Qwen3 auto device mapping cudaErrorAssert are unrelated." + "reason": "T5 tokenizer conflict and Qwen3 auto device mapping cuda assert are unrelated." }, { "left": "issue:43531", "right": "issue:44877", "accept": false, - "reason": "Sliding_window behavior in Qwen3-MoE and strict config preventing granite_speech loading are different problems." + "reason": "Sliding-window behavior in Qwen3-MoE and strict-config loading for granite_speech are different code paths." }, { "left": "issue:43425", "right": "issue:44403", "accept": false, - "reason": "Torch 2.10 incompatibility and loading-noise cleanup are not the same underlying issue." + "reason": "Torch version incompatibility and loading noise are unrelated." }, { "left": "issue:43611", "right": "issue:44534", "accept": false, - "reason": "base_model_prefix loading breakage and junk-filled non-persistent buffers are separate v5 regressions." + "reason": "`base_model_prefix` loading breakage and non-persistent buffer junk are separate v5 regressions." }, { "left": "issue:43632", "right": "issue:44534", "accept": false, - "reason": "_is_hf_initialized flag issues are unrelated to non-persistent buffer initialization." + "reason": "Different regressions: `_is_hf_initialized` flag handling vs non-persistent buffer serialization." }, { - "left": "issue:42175", - "right": "issue:45542", + "left": "issue:43792", + "right": "issue:44589", "accept": false, - "reason": "Missing TensorFlow in torch extra and tensorboard installed without TensorFlow are related packaging topics but not the same bug." + "reason": "Whisper runtime failure and missing Float8 storage object are unrelated." }, { - "left": "issue:43792", - "right": "issue:44589", + "left": "issue:42175", + "right": "issue:45542", "accept": false, - "reason": "Whisper inference failure and missing Float8 storage object are unrelated." + "reason": "Both mention TensorFlow/backends, but one is a packaging-extra complaint and the other is a runtime undefined-backend error; not clearly the same bug." }, { "left": "issue:40990", "right": "issue:44811", "accept": false, - "reason": "High perplexity on gpt-oss and Whisper batch_decode skip_special_tokens behavior are different model/runtime issues." + "reason": "Perplexity regression on GPT-OSS and Whisper batch_decode skip_special_tokens behavior are unrelated." }, { "left": "issue:44683", "right": "issue:45020", "accept": false, - "reason": "Compiled flex_attention on torch >=2.9 and remote_code loading regressions are unrelated." + "reason": "Compiled flex_attention on torch>=2.9 and remote_code loading regressions are different failures." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is well grounded in the packet and stays conservative on the soft edges. The only clearly duplicate-looking pair is indeed the Qwen3.5 save_pretrained regression (45216/45357), and the rest of the false verdicts are phrased cautiously enough given the titles. No overstatement stands out." + "feedback": "Grounded overall: the main v5 custom-init/loading cluster is supported by the titles, and the two accepted subclusters (45216/45357 and 44479/45325) look like the same bugfix family rather than separate issues. The soft-edge rejections are conservative and don\u2019t overclaim identity from similarity alone." }, "evaluator_used": true, "retried": false, @@ -3192,8 +3257,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -3252,11 +3317,11 @@ "issue:41553|issue:43582", "issue:43421|issue:44568", "issue:33666|issue:40444", - "issue:43818|issue:45357", "issue:44462|issue:45812", - "issue:43317|issue:43856", + "issue:43818|issue:45357", "issue:44829|issue:45479", "issue:43482|issue:45538", + "issue:43317|issue:43856", "issue:35707|issue:40444", "issue:44291|issue:44589", "issue:43381|issue:45923", @@ -3277,218 +3342,215 @@ "issue:43994|issue:44898", "issue:45325|issue:45834" ], - "prepared_review_unit_hash": "94293b722ee8bf4af78456f2865dc2bb7b42f5929793d8b8f7c349105e040480", + "prepared_review_unit_hash": "643939d9cdc5ba5b4775cd1ea5dbae9feec9752c7c57cb4c2f4177bc1feb4ef0", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36605, - "estimated_input_tokens": 9152, - "estimated_eval_tokens": 18560 + "serialized_chars": 36602, + "estimated_input_tokens": 9151, + "estimated_eval_tokens": 18558 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The items are mostly unrelated one-off issues, with only two soft edges looking like true duplicates: tokenizer special-token/post-processor handling, and a Qwen2.5-VL position-id regression. Everything else is too model- or symptom-specific to merge.", - "confidence": 0.78, - "canonical_issue_reason": "No single canonical issue for the whole set; the clearest duplicate cluster is issue 43421, which captures the underlying tokenizer special-token/post-processor regression more generally than its paired symptom report.", + "summary": "This cluster is heterogeneous: most items are distinct Transformers v5 regressions across loading, tokenization, multimodal position IDs, and training/runtime edge cases. No soft pair looks like a true duplicate, so all soft edges are rejected.", + "confidence": 0.74, + "canonical_issue_reason": "If one issue must anchor the cluster, 43611 is the closest umbrella: it\u2019s a broad Transformers v5 model-loading/init regression and is nearer to the other load/init failures than the specialized tokenizer or multimodal bugs.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43421 is the best issue anchor because it states the root tokenizer backend problem in a broad, actionable way and cleanly subsumes the paired symptom in 44568.", + "best_issue_reason": "43611 is the best representative of the dominant theme in this set: a generic v5 loading/init regression affecting model creation, which is the broadest and most central issue here.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44079", "right": "issue:45663", "accept": false, - "reason": "Different bugs: ModelOutput key assignment vs Gemma-4 FSDP2 shared_kv_states rebuilding." + "reason": "Different bugs: ModelOutput key assignment vs Gemma-4 FSDP2 shared_kv_states rebuild in attention." }, { "left": "issue:44403", "right": "issue:44991", "accept": false, - "reason": "Both involve loading, but one is generic noise on load and the other is tokenizer loading for a specific model; different root causes." + "reason": "Unrelated failures: noisy loading logs vs tokenizer loading regression for EMBEDDIA/est-roberta." }, { "left": "issue:41553", "right": "issue:43582", "accept": false, - "reason": "Bad tokenizer error message vs Apple Silicon allocator TypeError; unrelated subsystems." + "reason": "No overlap in code path: bad AutoTokenizer error messaging vs Apple Silicon caching_allocator_warmup TypeError." }, { "left": "issue:43421", "right": "issue:44568", - "accept": true, - "reason": "Both point to tokenizer special-token/post-processor handling regressing in v5; same underlying tokenizer backend bug." + "accept": false, + "reason": "Both involve special tokens, but one is runtime post-processor sync and the other is add_special_tokens not inserting BOS/EOS; not the same bug." }, { "left": "issue:33666", "right": "issue:40444", "accept": false, - "reason": "Both are Qwen-VL training issues, but one is multi-GPU training and the other is iterable dataset/multi-image finetuning; not the same bug." - }, - { - "left": "issue:43818", - "right": "issue:45357", - "accept": false, - "reason": "Video-LLaVA missing temporal attention/weight tying vs Qwen3.5 save_pretrained key regression; different code paths." + "reason": "Different Qwen2-VL/Qwen2.5-VL training failures with different triggers and inputs; not a duplicate." }, { "left": "issue:44462", "right": "issue:45812", "accept": false, - "reason": "Both touch AutoTokenizer, but one ignores tokenizer.json while the other yields wrong Granite token IDs; not enough evidence of one root cause." + "reason": "Both are AutoTokenizer regressions, but one ignores tokenizer.json while the other produces wrong token IDs for Granite models." }, { - "left": "issue:43317", - "right": "issue:43856", + "left": "issue:43818", + "right": "issue:45357", "accept": false, - "reason": "Dequantized model offload failure vs Qwen3 MoE memory efficiency; unrelated." + "reason": "Different model-family bugs: Video-LLaVA tower weights/temporal attention vs Qwen3.5 save_pretrained key names." }, { "left": "issue:44829", "right": "issue:45479", "accept": false, - "reason": "Different training failures: flash_attention_3 degeneracy vs num_labels=1 zero-loss classification behavior." + "reason": "Both are degenerate sequence-classification training reports, but one is tied to flash_attention_3 and the other to num_labels=1/problem_type handling." }, { "left": "issue:43482", "right": "issue:45538", "accept": false, - "reason": "Qwen2.5-GGUF loading failure vs CLIPTokenizer max-length constant bug; unrelated." + "reason": "Different areas entirely: Qwen2.5-GGUF loading vs CLIPTokenizer model_max_length handling." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Load/offload failure on device_map=auto vs Qwen3 MoE training memory inefficiency; not the same concrete bug." }, { "left": "issue:35707", "right": "issue:40444", "accept": false, - "reason": "Progressive generation with inputs_embeds/past_key_values vs Qwen2.5-VL multi-image iterable dataset finetuning; different bugs." + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL multi-image IterableDataset failures." }, { "left": "issue:44291", "right": "issue:44589", "accept": false, - "reason": "Both are v5 loading errors, but one is unexpected _is_hf_initialized argument and the other is missing Float8 storage; different failures." + "reason": "Both are loading-related v5 regressions, but one is an unexpected _is_hf_initialized argument and the other is a missing Float8 storage object." }, { "left": "issue:43381", "right": "issue:45923", "accept": false, - "reason": "Gradient checkpointing in eval mode is a usage/runtime issue, while the other is a missing capability flag on a trust_remote_code model." + "reason": "Gradient checkpointing eval-mode behavior is distinct from a missing supports_gradient_checkpointing flag on a trust_remote_code variant." }, { "left": "issue:43482", "right": "issue:43525", "accept": false, - "reason": "Qwen2.5-GGUF loading failure and Llama4Config missing pad_token_id are unrelated model-specific regressions." + "reason": "Different model/config issues: Qwen2.5-GGUF load failure vs Llama4Config missing pad_token_id." }, { "left": "issue:44079", "right": "issue:44534", "accept": false, - "reason": "ModelOutput key ordering and non-persistent buffer initialization are different v5 regressions." + "reason": "Different internal regressions: ModelOutput key assignment vs non-persistent buffer initialization." }, { "left": "issue:43377", "right": "issue:45491", "accept": false, - "reason": "Missing padding-mask support in MIMI encoder vs Gemma3 sliding-window all-padding NaNs; different architectures and failure modes." + "reason": "Both are batching/masking problems, but they affect different models and different mechanisms (padding mask support vs sliding-window all-padding windows)." }, { "left": "issue:43531", "right": "issue:44291", "accept": false, - "reason": "Sliding_window behavior in Qwen3-MoE vs init_empty_weights/_is_hf_initialized error; not the same issue." + "reason": "Qwen3-MoE sliding_window behavior is unrelated to init_empty_weights/_is_hf_initialized loading errors." }, { "left": "issue:41720", "right": "issue:45081", "accept": false, - "reason": "Qwen3 device-map cuda assert vs Mistral tokenizer regex patch crash; unrelated." + "reason": "Different subsystems: Qwen3 auto device mapping on A800 vs Mistral tokenizer regex patch crash." }, { "left": "issue:43749", "right": "issue:44079", "accept": false, - "reason": "FSDP CPU RAM efficient loading vs ModelOutput key assignment; unrelated." + "reason": "FSDP CPU RAM-efficient loading is a separate loading path from ModelOutput key assignment." }, { "left": "issue:40444", "right": "issue:45381", "accept": false, - "reason": "Multi-image iterable finetuning failure vs Qwen2.5-VL video vision_position_ids regression; different symptoms and likely different causes." + "reason": "Both are Qwen2.5-VL multimodal issues, but one is multi-image training input handling and the other is video vision_position_ids." }, { "left": "issue:43644", "right": "issue:44079", "accept": false, - "reason": "Non-persistent buffer junk on v5 vs ModelOutput key assignment; unrelated." + "reason": "Distinct v5 regressions: non-persistent buffers being filled with junk is not the same as ModelOutput key assignment." }, { "left": "issue:44265", "right": "issue:45072", "accept": false, - "reason": "torch.export with torch_compilable_check vs bfloat16 dtype mismatch in SwitchTransformers/TimmWrapperModel; different bugs." + "reason": "torch.export exportability issues are unrelated to bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." }, { "left": "issue:43632", "right": "issue:45072", "accept": false, - "reason": "_is_hf_initialized breakage vs bfloat16 dtype mismatch; not the same code-path problem." + "reason": "_is_hf_initialized breakage is a loading/init regression, while 45072 is a dtype mismatch during inference." }, { "left": "issue:42175", "right": "issue:43296", "accept": false, - "reason": "Packaging/backend dependency issue vs model loading failure in vLLM; unrelated." + "reason": "Packaging/dependency omission in pip install is unrelated to the PaddleOCR-VL load failure in vLLM." }, { "left": "issue:45325", "right": "issue:45381", - "accept": true, - "reason": "Both are Qwen2.5-VL position-id regressions in v5.3.0 and appear to stem from the same rope/indexing logic." + "accept": false, + "reason": "Related Qwen2.5-VL position-id bugs, but one is still-image temporal scaling and the other is video vision_position_ids; not the same concrete failure." }, { "left": "issue:43611", "right": "issue:43646", "accept": false, - "reason": "Both are transformers v5 breakages, but one is base_model_prefix loading and the other is custom model initialization; too broad to merge." + "reason": "Both are v5 init/loading regressions, but one is base_model_prefix loading and the other custom model initialization; too different to merge safely." }, { "left": "issue:43792", "right": "issue:43957", "accept": false, - "reason": "Whisper runtime failure vs meta-device loading regressions on specific models; unrelated." + "reason": "Whisper inference failure and torch.device(\"meta\") loading breakage are different bugs." }, { "left": "issue:43994", "right": "issue:44898", "accept": false, - "reason": "Different vision-model bugs: nonsensical SigLIP2 outputs vs Perceiver non-default resolution failure." + "reason": "Different model families and failure modes: SigLIP2 nonsense outputs vs Perceiver non-default resolution failure." }, { "left": "issue:45325", "right": "issue:45834", "accept": false, - "reason": "Position-id regression in Qwen2.5-VL vs Kosmos2.5 OCR index error; different model families and likely different fixes." + "reason": "Both touch Qwen2.5-VL, but one is a specific rope-index scaling bug and the other is a long-OCR index error; not enough evidence of the same bug." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The analysis is grounded in the packet and stays conservative overall. The two accepted soft pairs are plausible duplicate/merge candidates from the titles alone, and the remaining pairs are reasonably rejected as too distinct. No obvious overstatement stands out." - }, - "evaluator_used": true, + "evaluator_result": null, + "evaluator_used": false, "retried": false, - "accepted_nontrivial_soft_edge": true, + "accepted_nontrivial_soft_edge": false, "error_kind": null, "error_message": null }, @@ -3500,8 +3562,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -3562,8 +3624,8 @@ "issue:44479|issue:45823", "issue:43701|issue:45325", "issue:44291|issue:44877", - "issue:43881|issue:44488", "issue:43606|issue:44265", + "issue:43881|issue:44488", "issue:43262|issue:43746", "issue:43873|issue:45701", "issue:44568|issue:45920", @@ -3571,8 +3633,8 @@ "issue:40990|issue:45584", "issue:44186|issue:44265", "issue:35707|issue:41720", - "issue:41720|issue:45823", "issue:43606|issue:43819", + "issue:41720|issue:45823", "issue:43454|issue:43550", "issue:43377|issue:44610", "issue:43317|issue:43873", @@ -3585,20 +3647,20 @@ "issue:40990|issue:42947", "issue:43526|issue:43883" ], - "prepared_review_unit_hash": "262303bc5773319632530055cf3a3d64ec822385e61e2adcb3698a53f9658998", + "prepared_review_unit_hash": "c9aba62c634d781729304c6dec86d10601d79cffb59d20725e399d5a0d8a0a49", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36522, + "serialized_chars": 36524, "estimated_input_tokens": 9131, "estimated_eval_tokens": 18518 }, @@ -3607,186 +3669,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is not a duplicate cluster: the issues span many unrelated subsystems (tokenization, offloading, quantization, multimodal processors, config loading, export, and training). No single canonical issue or PR emerges from the titles alone.", + "summary": "The cluster is heterogeneous: most pairs share only broad subsystem overlap (quantization, offloading, multimodal, tokenization, audio) but describe different concrete failures and code paths. No soft pair looks like a true duplicate or mergeable same-fix PR case.", "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue fits the cluster; the items describe unrelated bugs and regressions rather than one underlying defect.", + "canonical_issue_reason": "issue:43873 is the broadest representative in the cluster: it\u2019s an open, actively discussed offloading/quantization bug with inbound references and sits near several nearby offload-related reports, even though the cluster overall is not a duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "No globally best issue candidate exists because none of the issues subsumes the others or represents the same concrete problem set.", + "best_issue_reason": "issue:43873 best captures the most central recurring theme in the cluster (offloading with quantization) and has the strongest evidence of broader impact and follow-up activity.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:42915", "right": "issue:43482", "accept": false, - "reason": "Different problems: FP8/DeepSpeed weight conversion vs a v5 GGUF loading failure." + "reason": "Different failures: Qwen3 MoE fine-grained FP8 config vs Qwen2.5-GGUF loading under transformers v5. Same broad model-loading area, but not the same bug or fix." }, { "left": "issue:43454", "right": "issue:44466", "accept": false, - "reason": "Both mention lm_head/tied weights, but one is missing weight tying and the other is inconsistent serialization behavior." + "reason": "Both mention tied weights/lm_head, but one is a runtime generation bug in AyaVisionConfig and the other is a serialization inconsistency in v5/main. Different code paths." }, { "left": "issue:43296", "right": "issue:45800", "accept": false, - "reason": "vLLM model-load failure vs a broad torch/transformers version incompatibility; not the same bug." + "reason": "PaddleOCR-VL load failure in vLLM/transformers 4.53.0 vs a torch 2.4.1 / transformers 5.8.0 incompatibility. Separate compatibility issues." }, { "left": "issue:44387", "right": "issue:44617", "accept": false, - "reason": "Both are OOM-related, but one is int4 reserved-memory growth and the other is a Sam3Video memory issue." + "reason": "Both are OOM-related, but one is int4 CUDA reserved-memory growth in Transformers 5.x and the other is Sam3Video model OOM. Too generic to be the same bug." }, { "left": "issue:44479", "right": "issue:45823", "accept": false, - "reason": "Different models and failure modes: video-input regression vs device mismatch in forward with device_map=auto." + "reason": "Both involve multimodal model/device behavior, but one is a video input regression across Qwen VL variants and the other is a PLE device mismatch during forward on Gemma4. Different failures." }, { "left": "issue:43701", "right": "issue:45325", "accept": false, - "reason": "Checkpoint key mismatch is unrelated to Qwen2.5-VL rope-index scaling." + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL rope index scaling. No shared code path." }, { "left": "issue:44291", "right": "issue:44877", "accept": false, - "reason": "Unexpected _is_hf_initialized argument from init_empty_weights is unrelated to strict granite_speech config loading." + "reason": "`init_empty_weights` unexpected argument error vs strict config blocking granite_speech config loading. Both are loading issues, but not the same defect." }, { - "left": "issue:43881", - "right": "issue:44488", + "left": "issue:43606", + "right": "issue:44265", "accept": false, - "reason": "Different load failures for different models; no shared concrete code path is evident." + "reason": "CPU offload device mismatch for suno/bark-small vs torch.export failing on `torch_compilable_check`. Different subsystems and fixes." }, { - "left": "issue:43606", - "right": "issue:44265", + "left": "issue:43881", + "right": "issue:44488", "accept": false, - "reason": "CPU-offload device mismatch is unrelated to torch.export failing on torch_compilable_check." + "reason": "Generic model loading failure for glm-4v-9b vs inability to load a specific BERT model. No evidence of the same underlying bug." }, { "left": "issue:43262", "right": "issue:43746", "accept": false, - "reason": "Audio chat-template sample-rate defaulting is unrelated to PEFT adapter local-checkpoint loading." + "reason": "Audio processor chat-template sample rate default vs GraniteSpeech PEFT local checkpoint loading. Different audio/model-loading issues." }, { "left": "issue:43873", "right": "issue:45701", "accept": false, - "reason": "Offloading/quantization behavior is unrelated to a tokenizer-version regression." + "reason": "Quantization/offloading behavior vs tokenizer-version-dependent tokenization changes. Overlap is only that both are Transformers regressions." }, { "left": "issue:44568", "right": "issue:45920", "accept": false, - "reason": "Specific BOS/EOS insertion failure for one tokenizer is not the same as broad wrong-token-ID regressions across models." + "reason": "mDeBERTa special-token handling vs broad AutoTokenizer wrong-token-ID regression across many models. Same area, but 45920 is a wider tokenizer regression, not the same concrete bug." }, { "left": "issue:34634", "right": "issue:35707", "accept": false, - "reason": "Bark voice_preset failure is unrelated to progressive generation with inputs_embeds and past_key_values." + "reason": "BarkProcessor voice_preset issue and progressive generation with inputs_embeds/past_key_values are unrelated bugs." }, { "left": "issue:40990", "right": "issue:45584", "accept": false, - "reason": "Perplexity regression on gpt-oss-20b is unrelated to Whisper empty-transcription failure." + "reason": "High perplexity on gpt-oss-20b is a model/perplexity issue; Whisper empty transcription after align_special_tokens is a generation edge case. Different code paths." }, { "left": "issue:44186", "right": "issue:44265", "accept": false, - "reason": "Tokenizer crashes on NER/padding are unrelated to torch.export and torch_compilable_check." + "reason": "LayoutLMv2Tokenizer crashes on NER/batching vs torch.export with torch_compilable_check. Both touch tooling, but not the same problem." }, { "left": "issue:35707", "right": "issue:41720", "accept": false, - "reason": "Progressive generation bug is unrelated to Qwen3 auto-device-map CUDA asserts." + "reason": "Progressive generation with inputs_embeds/past_key_values vs Qwen3 auto device mapping cudaErrorAssert. No meaningful overlap beyond generation/runtime failures." }, { - "left": "issue:41720", - "right": "issue:45823", + "left": "issue:43606", + "right": "issue:43819", "accept": false, - "reason": "Both involve device_map=auto, but the affected models and underlying failures differ." + "reason": "Bark CPU offload device mismatch vs DAC.from_latents mismatch due to missing STE. Both are model-forward inconsistencies, but different models and root causes." }, { - "left": "issue:43606", - "right": "issue:43819", + "left": "issue:41720", + "right": "issue:45823", "accept": false, - "reason": "CPU-offload device mismatch for Bark is unrelated to DAC.from_latents not matching forward without STE." + "reason": "Qwen3 auto device-map cudaErrorAssert vs Gemma4 PLE device mismatch. Both are device-placement failures, but not the same code path or model family." }, { "left": "issue:43454", "right": "issue:43550", "accept": false, - "reason": "Weight-tying bug in AyaVisionConfig is unrelated to Bamba SDPA/torch.compile failure." + "reason": "AyaVision tied lm_head weights vs Bamba-9B-v2 torch.compile/SDPA failure. Different model internals and failure modes." }, { "left": "issue:43377", "right": "issue:44610", "accept": false, - "reason": "MIMI batching/padding-mask mismatch is unrelated to OmDet-Turbo processor image-size mismatch." + "reason": "MIMI encoder batching/padding-mask bug vs OmDet-Turbo processor/model input-size mismatch. Both are processor/model alignment issues, but not the same defect." }, { "left": "issue:43317", "right": "issue:43873", "accept": false, - "reason": "Dequantized-model load failure on gpu+cpu offload is a different bug from generic quantization offloading behavior." + "reason": "Dequantized model load failure with gpu+cpu offload is adjacent to offloading/quantization, but the reports describe different behaviors and scenarios; not clearly the same underlying bug." }, { "left": "issue:39692", "right": "issue:40444", "accept": false, - "reason": "SigLIP2 doc/example mistakes are unrelated to Qwen2.5-VL iterable-dataset multi-image finetuning failures." + "reason": "SigLIP2 docs example errors vs Qwen2.5-VL finetuning with iterable multi-image datasets. Both are multimodal, but unrelated concrete bugs." }, { "left": "issue:43454", "right": "issue:45072", "accept": false, - "reason": "lm_head tying/serialization issues are unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel." + "reason": "AyaVision lm_head tying bug vs bfloat16 dtype mismatches in CI for SwitchTransformers/TimmWrapperModel. Different issues." }, { "left": "issue:43572", "right": "issue:45759", "accept": false, - "reason": "Missing pad_token_idx in StableLmConfig is unrelated to not unwrapping text_config in composite Qwen multimodal configs." + "reason": "Missing pad_token_idx in StableLmConfig vs AutoModelForCausalLM.from_config not unwrapping text_config for composite Qwen multimodal configs. Both are config-related, but not the same bug." }, { "left": "issue:43377", "right": "issue:44898", "accept": false, - "reason": "Batched-vs-single encoder output mismatch is unrelated to Perceiver non-default-resolution failures." + "reason": "MIMI padding-mask batch discrepancy vs Perceiver image classification failing at non-default resolution. Different model families and failures." }, { "left": "issue:43257", "right": "issue:45923", "accept": false, - "reason": "MOE weight conversion with accelerate+deepspeed is unrelated to a missing supports_gradient_checkpointing flag." + "reason": "Qwen3 MoE loading with accelerate+deepspeed vs missing supports_gradient_checkpointing on a trust_remote_code variant. Both involve Qwen/Nemotron model plumbing, but not the same issue." }, { "left": "issue:43322", "right": "issue:46050", "accept": false, - "reason": "Llava Next load segfault is unrelated to quantized model save/load breakage after sinq." + "reason": "Llava Next segmentation fault on load vs sinq quantized model save/load breakage. Different failure class and lifecycle stage." }, { "left": "issue:40990", "right": "issue:42947", "accept": false, - "reason": "High perplexity on a language-model benchmark is unrelated to gradient checkpointing being ineffective with PEFT LoRA." + "reason": "High perplexity evaluation issue vs gradient checkpointing ineffective with PEFT LoRA. No shared bug mechanism." }, { "left": "issue:43526", "right": "issue:43883", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels is unrelated to the missing all_tied_weights_keys attribute on MolmoForCausalLM." + "reason": "BeitImageProcessorFast reduce_labels bug vs Molmo missing all_tied_weights_keys AttributeError. Unrelated processor vs model attribute issue." } ] }, @@ -3805,8 +3867,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -3862,9 +3924,9 @@ "soft_pairs": [ "issue:43454|issue:43720", "issue:44466|issue:45357", - "issue:43749|issue:44534", - "issue:41720|issue:43322", "issue:43296|issue:44877", + "issue:41720|issue:43322", + "issue:43749|issue:44534", "issue:43881|issue:44991", "issue:41720|issue:44977", "issue:42617|issue:42915", @@ -3874,12 +3936,12 @@ "issue:44186|issue:45072", "issue:43577|issue:43708", "issue:44387|issue:45704", - "issue:36331|issue:42617", "issue:44877|issue:44991", + "issue:36331|issue:42617", "issue:43873|issue:44756", "issue:36331|issue:42915", - "issue:45632|issue:45684", "issue:44423|issue:45406", + "issue:45632|issue:45684", "issue:43450|issue:45701", "issue:43550|issue:43720", "issue:41720|issue:42617", @@ -3894,232 +3956,232 @@ "issue:43526|issue:43881", "issue:45081|issue:45812" ], - "prepared_review_unit_hash": "e2179cecee1fcdd6785bf83def2259ff14efc2eda03a62cb746132b749984f61", + "prepared_review_unit_hash": "122c752f4c223b2f07b7de5d0102ff91fe3a89bdee350a7f64293d586cecf754", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37820, - "estimated_input_tokens": 9455, - "estimated_eval_tokens": 19166 + "serialized_chars": 37813, + "estimated_input_tokens": 9454, + "estimated_eval_tokens": 19164 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are a loose collection of unrelated transformer/model bugs. None of the soft pairs look like the same underlying issue, so there is no convincing duplicate canonical across the set.", - "confidence": 0.97, - "canonical_issue_reason": null, + "summary": "The cluster is a loose set of mostly unrelated Transformers issues; despite some superficial keyword overlap around loading, offloading, tokenizers, and multimodal models, none of the soft pairs look like true duplicates. I\u2019d anchor the cluster on the broad offloading/quantization report.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43873 is the broadest and most central report in the set: it is open, has the most discussion/inbound references, and covers a general offloading/quantization failure mode that can absorb nearby loading/offload regressions.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43873 is the strongest standalone report: clear reproduction, active discussion/inbound references, and a broad offloading/quantization regression. It is still not a duplicate of the other issues, just the most representative single bug report in the batch.", + "best_issue_reason": "issue:43873 is the best representative issue for the cluster because it is broad, active, and likely to attract related offloading/quantization follow-ups.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43454", "right": "issue:43720", "accept": false, - "reason": "Different bugs: one is tied lm_head weights in Mistral/AyaVision-style generation, the other is BitNet packed-weight unpacking during accelerate loading." + "reason": "Different bugs: tied lm_head weights in AyaVision vs unpacking packed BitNet weights during accelerate loading." }, { "left": "issue:44466", "right": "issue:45357", "accept": false, - "reason": "Both mention weights/serialization, but one is lm_head serialization by device and the other is incorrect visual encoder keys in Qwen3.5 save_pretrained." + "reason": "Both mention lm_head/serialization, but one is a device-dependent save_pretrained inconsistency and the other is incorrect visual encoder keys in Qwen3.5." }, { - "left": "issue:43749", - "right": "issue:44534", + "left": "issue:43296", + "right": "issue:44877", "accept": false, - "reason": "Different regressions: FSDP CPU RAM efficient loading vs non-persistent buffers being filled with junk in v5." + "reason": "Different failure surfaces: PaddleOCR-VL load failure in vLLM vs strict config blocking granite_speech config loading." }, { "left": "issue:41720", "right": "issue:43322", "accept": false, - "reason": "Different failure modes and code paths: auto device mapping on Qwen3 vs Llava Next loading segfault." + "reason": "Different symptoms and models: cuda assert with Qwen3 auto device mapping vs Llava Next segmentation fault." }, { - "left": "issue:43296", - "right": "issue:44877", + "left": "issue:43749", + "right": "issue:44534", "accept": false, - "reason": "Unrelated models and root causes: vLLM/PaddleOCR-VL load failure vs strict config blocking granite_speech." + "reason": "Both are loading-related, but one is an FSDP CPU RAM efficient loading regression and the other is a v5 non-persistent buffer corruption issue." }, { "left": "issue:43881", "right": "issue:44991", "accept": false, - "reason": "Different subsystems: GLM-4V-9B model loading failure vs tokenizer loading regression for est-roberta." + "reason": "Different components: glm-4v-9b model loading vs tokenizer loading for EMBEDDIA/est-roberta." }, { "left": "issue:41720", "right": "issue:44977", "accept": false, - "reason": "Same family name but not the same bug: auto device mapping assert vs flash-attention generation regression." + "reason": "Same model family, but different concrete bug: device-map CUDA assert vs flash-attention generation regression." }, { "left": "issue:42617", "right": "issue:42915", "accept": false, - "reason": "Different problems: 3d_parallel.py execution failure vs Qwen3Moe failure with FineGrainedFP8Config." + "reason": "Unrelated issues: a 3d_parallel.py runtime failure vs Qwen3Moe failing with FineGrainedFP8Config." }, { "left": "issue:44008", "right": "issue:45823", "accept": false, - "reason": "Both Gemma-related, but one is a variable name collision in forward() and the other is a device mismatch in PLE forwarding." + "reason": "Both are Gemma-related, but one is a variable-name collision in forward() and the other is a device-mismatch in PLE forwarding." }, { "left": "issue:40990", "right": "issue:45701", "accept": false, - "reason": "High perplexity on a model is not the same as a tokenization version regression, even if tokenization could influence metrics." + "reason": "High perplexity on gpt-oss-20b is not the same as a transformers-version tokenization change." }, { "left": "issue:43404", "right": "issue:45357", "accept": false, - "reason": "Different models and defects: lm_head tying in Mistral3ForConditionalGeneration vs Qwen3.5 visual encoder key serialization." + "reason": "Both concern model weights, but one is untied lm_head weights in Mistral3 and the other is incorrect visual encoder keys on save_pretrained." }, { "left": "issue:44186", "right": "issue:45072", "accept": false, - "reason": "Tokenizer crash on NER/padding is unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel inference." + "reason": "Tokenizer crashes on NER/padding are unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel inference." }, { "left": "issue:43577", "right": "issue:43708", "accept": false, - "reason": "Different layers of the stack: BLIP2 dtype propagation vs Trainer resume/max_steps calculation." + "reason": "Different areas: BLIP2 dtype propagation vs trainer resume_from_checkpoint max_steps calculation." }, { "left": "issue:44387", "right": "issue:45704", "accept": false, - "reason": "Both memory-related, but one is int4 CUDA reserved memory/OOM in Transformers 5.x and the other is an apex RMSNorm leak in T5." + "reason": "Distinct memory problems: int4 quantization reserved-memory OOM vs apex FusedRMSNorm memory leak." }, { - "left": "issue:36331", - "right": "issue:42617", + "left": "issue:44877", + "right": "issue:44991", "accept": false, - "reason": "Completely different errors: custom compute_loss signature mismatch vs a script execution failure." + "reason": "Different loading paths: strict config rejection vs tokenizer failure for a specific Roberta model." }, { - "left": "issue:44877", - "right": "issue:44991", + "left": "issue:36331", + "right": "issue:42617", "accept": false, - "reason": "Different loading paths: strict config validation for granite_speech vs tokenizer loading regression for est-roberta." + "reason": "Training API signature error vs 3d_parallel.py execution failure are unrelated." }, { "left": "issue:43873", "right": "issue:44756", "accept": false, - "reason": "Both mention loading/offload behavior, but one is quantization offloading and the other is mmap/OOM on Strix Halo." + "reason": "Both mention memory/offload, but one is quantization offloading broken and the other is mmap causing OOM on Strix Halo." }, { "left": "issue:36331", "right": "issue:42915", "accept": false, - "reason": "Unrelated: Trainer compute_loss API mismatch vs Qwen3Moe failing with FineGrainedFP8Config." + "reason": "Different problems: custom trainer compute_loss signature mismatch vs Qwen3Moe fine-grained FP8 loading failure." }, { - "left": "issue:45632", - "right": "issue:45684", + "left": "issue:44423", + "right": "issue:45406", "accept": false, - "reason": "Different save/load issues: trust_remote_code cache path collision vs read-only permission propagation during save_pretrained." + "reason": "Both affect serve, but one is a continuous-batching multimodal crash and the other is a Gemma4Processor missing _tokenizer attribute." }, { - "left": "issue:44423", - "right": "issue:45406", + "left": "issue:45632", + "right": "issue:45684", "accept": false, - "reason": "Both are serve-time multimodal crashes, but the attribute errors and processor paths differ; not the same concrete bug." + "reason": "Unrelated save/cache bugs: trust_remote_code cache-path collision vs read-only permission propagation in save_pretrained." }, { "left": "issue:43450", "right": "issue:45701", "accept": false, - "reason": "Video processor batching shape bug is unrelated to tokenizer-version-dependent tokenization changes." + "reason": "Video batch-shape regression is unrelated to transformers-version tokenization differences." }, { "left": "issue:43550", "right": "issue:43720", "accept": false, - "reason": "Different causes: torch.compile/SDPA failure in Bamba vs packed-weight loading failure in BitNet." + "reason": "Both are model loading/execution issues, but one is torch.compile+SDPA on Bamba and the other is BitNet packed-weight loading." }, { "left": "issue:41720", "right": "issue:42617", "accept": false, - "reason": "Different models and failures: Qwen3 auto device mapping assert vs 3d_parallel.py not running." + "reason": "Different failures: Qwen3 device-map CUDA assert vs inability to run 3d_parallel.py." }, { "left": "issue:44479", "right": "issue:44918", "accept": false, - "reason": "Both Qwen3.5-adjacent, but one is a video-input regression and the other is an SFT trainer embedding unpacking issue." + "reason": "Both involve Qwen3.5, but one is a video-input regression and the other is TRL SFT embedding unpacking." }, { "left": "issue:44610", "right": "issue:44898", "accept": false, - "reason": "Different vision bugs: OmDet-Turbo processor output size mismatch vs Perceiver interpolation failure on non-default resolution." + "reason": "Both are image size/preprocessing complaints, but they involve different models and different expected resolutions/mechanisms." }, { "left": "issue:44186", "right": "issue:44610", "accept": false, - "reason": "Tokenizer/NER padding crash is unrelated to image processor resolution mismatch." + "reason": "Tokenizer padding/truncation crashes are not the same bug as OmDet-Turbo producing the wrong image size." }, { "left": "issue:43550", "right": "issue:43606", "accept": false, - "reason": "Different runtime paths: SDPA under torch.compile vs CPU offload device mismatch." + "reason": "Different device/runtime bugs: torch.compile with SDPA vs CPU-offload device mismatch in bark-small." }, { "left": "issue:42915", "right": "issue:43317", "accept": false, - "reason": "Both involve model loading/offload, but the underlying failures are different: FineGrainedFP8Config vs dequantized model loading on gpu+cpu offload." + "reason": "Different loading failures: Qwen3Moe FineGrainedFP8Config vs dequantized model load with gpu+cpu offload." }, { "left": "issue:43295", "right": "issue:45701", "accept": false, - "reason": "Regression in custom model image/text processor handling vs a tokenizer behavior change across versions." + "reason": "Regression in processor/tokenizer image handling is not the same as a version-dependent tokenization change." }, { "left": "issue:41720", "right": "issue:45161", "accept": false, - "reason": "Different Qwen/GPT-OSS problems: auto device mapping crash vs tensor-parallel-only failure." + "reason": "Different Qwen-OSS issues: auto device mapping on A800 vs tensor-parallel not working for GPT-OSS MoE." }, { "left": "issue:44671", "right": "issue:44743", "accept": false, - "reason": "Different model families and bugs: CamemBERT masked LM predictions vs Qwen3.5 recurrent state reset with cache." + "reason": "Separate model bugs: CamemBERT masked LM predictions vs Qwen3.5 recurrent state reset with cache." }, { "left": "issue:43526", "right": "issue:43881", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug is unrelated to GLM-4V-9B loading failure." + "reason": "Unrelated processors/models: BeitImageProcessorFast reduce_labels bug vs glm-4v-9b loading failure." }, { "left": "issue:45081", "right": "issue:45812", "accept": false, - "reason": "Both tokenizer-related, but one is a Mistral regex patch crash and the other is wrong token IDs for Granite models; different code paths." + "reason": "Both are tokenizer-related, but they affect different models and different failure modes; not the same underlying bug." } ] }, @@ -4138,8 +4200,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -4222,211 +4284,205 @@ "issue:44451|issue:44991", "issue:44898|issue:45072" ], - "prepared_review_unit_hash": "fb6313a4a473db6046ca2ad5697b92da4c20354e088abdc2fa0a8a1e1e7179d3", + "prepared_review_unit_hash": "18616d84faa2811affa9e5e4159c6d91429f9c401f3286141fb0d8972bef3474", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36164, - "estimated_input_tokens": 9041, - "estimated_eval_tokens": 18338 + "serialized_chars": 36173, + "estimated_input_tokens": 9044, + "estimated_eval_tokens": 18344 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most pairs are thematically related only at a broad subsystem level (tokenizers, multimodal processors, loading/offloading), but only one pair looks like a true duplicate of the same concrete bug: the two Qwen2_5_VLProcessor.apply_chat_template crash reports.", - "confidence": 0.83, - "canonical_issue_reason": "Issue 44514 is the earlier report of the exact same crash (`Qwen2_5_VLProcessor.apply_chat_template` on batched input with `padding=False`), so it is the best canonical issue for the duplicate pair.", + "summary": "The cluster is mostly a set of unrelated, model-specific issues. The only clear duplicate pair is issue:44514 and issue:44545, which have nearly identical titles and point to the same Qwen2_5_VLProcessor batched-input crash surface.", + "confidence": 0.86, + "canonical_issue_reason": "issue:44514 is the best canonical issue candidate based on the visible metadata: it has the clearer, more specific title for the shared Qwen2_5_VLProcessor.apply_chat_template batched-input crash when padding=False, and issue:44545 appears to be the same report in near-verbatim form.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44514 is the strongest representative because it is the first and most complete report of the shared batched-input `apply_chat_template` failure.", + "best_issue_reason": "issue:44514 is the strongest global issue representative because it most clearly names the concrete failure mode and subsystem, and it is the closest match to its duplicate counterpart.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43531", "right": "issue:43572", "accept": false, - "reason": "Both are model/config bugs, but one is a Qwen3-MoE sliding-window issue and the other is a StableLmConfig padding-token regression; different failures and code paths." + "reason": "Both are tokenizer/config bugs, but they involve different models and different fields (sliding_window vs pad_token_idx), so they are not the same defect." }, { "left": "issue:36010", "right": "issue:42617", "accept": false, - "reason": "An import error for GenerationMixin is unrelated to a 3d_parallel.py runtime failure." + "reason": "GenerationMixin import failure and a 3d_parallel.py execution problem are different failure modes with no clear shared code path." }, { "left": "issue:43316", "right": "issue:44913", "accept": false, - "reason": "Different config problems: Gemma3TextConfig API inconsistency vs GPTNeoX rotary_pct not persisting on reload." + "reason": "Both are config inconsistencies, but they affect different models and different properties, so they do not look like one underlying bug." }, { "left": "issue:43540", "right": "issue:43782", "accept": false, - "reason": "Both involve Qwen multimodal models, but one is video-input processing and the other is from_pretrained weight_only loading." + "reason": "These are both Qwen multimodal issues, but one is a video-input ValueError and the other is a from_pretrained weight_only error; different bug surfaces." }, { "left": "issue:42915", "right": "issue:43322", "accept": false, - "reason": "Qwen3Moe FP8 failure and Llava Next segfault are unrelated model-loading/runtime crashes." + "reason": "Qwen3Moe FP8 configuration failure and Llava Next segmentation fault are unrelated model-loading/runtime issues." }, { "left": "issue:43421", "right": "issue:45812", "accept": false, - "reason": "Runtime post-processor refresh for special tokens is a different bug from Granite token IDs being wrong on AutoTokenizer load." + "reason": "Both involve tokenization, but one is runtime post-processor updates after special-token changes and the other is a tokenizer regression for Granite models; not the same concrete bug." }, { "left": "issue:44186", "right": "issue:44361", "accept": false, - "reason": "Both are tokenizer/task crashes, but they affect different tokenizers and different failure modes." + "reason": "Tokenizer crashes are too broad here; these affect different tokenizers and task flows without evidence of a shared code path." }, { "left": "issue:43388", "right": "issue:45059", "accept": false, - "reason": "A gather_for_metrics tuple-label batching bug is not the same as SAM3 text/bbox behavior." + "reason": "Metrics label handling and SAM3 text/bounding-box behavior are unrelated problems." }, { "left": "issue:29159", "right": "issue:36032", "accept": false, - "reason": "Generic slow/fast tokenizer inconsistency is not the same as the T5 add_special_tokens name conflict on load." + "reason": "A slow-vs-fast tokenizer inconsistency is not the same issue as the T5Tokenizer add_special_tokens AttributeError." }, { "left": "issue:45406", "right": "issue:45464", "accept": false, - "reason": "Both touch serving, but one is a missing _tokenizer attribute in Gemma4Processor and the other is a Qwen3.5 streaming API failure." + "reason": "These are different API paths: a serve-side processor attribute error versus a streaming chat/completions failure." }, { "left": "issue:33357", "right": "issue:35707", "accept": false, - "reason": "A MacOS CLIP bus error is unrelated to progressive generation with inputs_embeds and past_key_values." + "reason": "A MacOS bus error in CLIP loading and progressive generation with inputs_embeds/past_key_values are unrelated." }, { "left": "issue:30064", "right": "issue:34634", "accept": false, - "reason": "Image processor void segmentation-map handling and BarkProcessor voice_preset are different features." + "reason": "Void segmentation-map processing and BarkProcessor voice_preset behavior are different subsystems and bugs." }, { "left": "issue:41720", "right": "issue:43746", "accept": false, - "reason": "Auto device mapping cudaErrorAssert and PEFT local-checkpoint loading are different problems." + "reason": "Auto device mapping cudaErrorAssert and PEFT local-checkpoint loading are different failures." }, { "left": "issue:43421", "right": "issue:45701", "accept": false, - "reason": "One is about runtime special-token post-processing, the other is a broad version-related tokenization change." + "reason": "A runtime tokenizer backend sync bug and a broad version-related tokenization regression are not clearly the same issue." }, { "left": "issue:43606", "right": "issue:43720", "accept": false, - "reason": "CPU-offload device mismatch and packed-weight unpacking during accelerate loading are distinct offloading bugs." + "reason": "CPU offload device mismatch and packed-weight unpacking during accelerate loading are distinct bugs." }, { "left": "issue:43550", "right": "issue:43653", "accept": false, - "reason": "A torch.compile SDPA failure is unrelated to a tokenizer special-token registration bug." + "reason": "torch.compile/SDPA failure in Bamba-9B-v2 is unrelated to BigBirdTokenizer special-token registration." }, { "left": "issue:44492", "right": "issue:44509", "accept": false, - "reason": "A typo in cache strategies is unrelated to outdated pipeline-task docs." + "reason": "A typo in cache-strategy docs and stale pipeline-task documentation are not the same problem." }, { "left": "issue:43873", "right": "issue:45704", "accept": false, - "reason": "Quantization/offloading behavior and an apex RMSNorm memory leak are different issues." + "reason": "Quantization/offloading behavior and a T5 apex FusedRMSNorm memory leak are different issues." }, { "left": "issue:43606", "right": "issue:43653", "accept": false, - "reason": "Device mismatch under CPU offload is not the same as BigBirdTokenizer empty decode output." + "reason": "Device mismatch during CPU offload and tokenizer special-token registration/decode output are unrelated." }, { "left": "issue:43531", "right": "issue:45759", "accept": false, - "reason": "Both mention Qwen-family models, but sliding_window behavior and AutoModelForCausalLM.from_config text_config unwrapping are different code paths." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to text_config unwrapping in composite Qwen multimodal configs." }, { "left": "issue:34634", "right": "issue:41720", "accept": false, - "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping have no underlying overlap." + "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping errors are different model-specific failures." }, { "left": "issue:42490", "right": "issue:43329", "accept": false, - "reason": "A trailing-slash import warning is unrelated to the multimodal-token counting bug in video processing." + "reason": "A local-path import warning and undefined variables in multimodal token counting are different bugs." }, { "left": "issue:44514", "right": "issue:44545", "accept": true, - "reason": "They have the same title and describe the same batched `Qwen2_5_VLProcessor.apply_chat_template` crash with `padding=False`." + "reason": "The titles are effectively the same and describe the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash when padding=False." }, { "left": "issue:34567", "right": "issue:35707", "accept": false, - "reason": "TrainerState token counting and progressive generation are different bugs." + "reason": "TrainerState token counting and progressive generation with inputs_embeds/past_key_values are unrelated." }, { "left": "issue:44625", "right": "issue:45479", "accept": false, - "reason": "Qwen3.5 num_labels propagation and the generic single-label-classification zero-loss bug are not the same defect." - }, - { - "left": "issue:43421", - "right": "issue:43927", - "accept": false, - "reason": "Runtime special-token refresh and DiaConfig losing custom token IDs after save/load affect different components and failure modes." + "reason": "Config propagation in Qwen3.5 and a degenerate zero-loss bug across classification models are not the same concrete defect." }, { "left": "issue:44451", "right": "issue:44991", "accept": false, - "reason": "Both are model-specific tokenizer load failures, but the titles do not indicate the same underlying regression." + "reason": "A model-loading failure for one checkpoint and a tokenizer-loading regression for another model are different issues." }, { "left": "issue:44898", "right": "issue:45072", "accept": false, - "reason": "Perceiver non-default-resolution failure and dtype mismatches in bfloat16 inference are unrelated." + "reason": "Perceiver interpolation failure and dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated model-specific bugs." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The canonical duplicate choice is well grounded: 44514 and 44545 have effectively the same title and describe the same batched `Qwen2_5_VLProcessor.apply_chat_template` crash. The non-matching pairs are mostly rejected conservatively based on distinct titles, models, or failure modes, without overclaiming equivalence." + "accept": false, + "feedback": "The duplicate call on issue:44514/44545 is well supported by the titles, but the reasoning slightly overreaches the packet by asserting that 44514 is \"reproducible\" and the clear canonical issue without body evidence. With only metadata/titles visible here, that should be stated more conservatively. The soft-edge rejections are mostly cautious, but the overall confidence and canonicality claim are a bit too strong for the evidence shown." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -4439,8 +4495,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -4499,10 +4555,10 @@ "issue:44077|issue:44625", "issue:45759|issue:45859", "issue:44918|issue:44936", - "issue:44368|issue:45357", "issue:44062|issue:44451", - "issue:39692|issue:44610", + "issue:44368|issue:45357", "issue:45588|issue:45759", + "issue:39692|issue:44610", "issue:45704|issue:45923", "issue:43746|issue:46050", "issue:43421|issue:43653", @@ -4518,26 +4574,26 @@ "issue:43596|issue:44220", "issue:43746|issue:43873", "issue:39692|issue:43873", - "issue:40444|issue:43746", "issue:45356|issue:45907", + "issue:40444|issue:43746", "issue:44220|issue:44661", "issue:30064|issue:36010", "issue:43720|issue:43819" ], - "prepared_review_unit_hash": "02decd14b55f765f091d4f8aee083dfd40a24eddeb9551832a604dc79a4466f9", + "prepared_review_unit_hash": "1ff08558b0fe80be4d5ee0fda49072cc508380642827cbcebd44802ba170d26c", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 30, - "serialized_chars": 36474, + "serialized_chars": 36473, "estimated_input_tokens": 9119, "estimated_eval_tokens": 18494 }, @@ -4546,192 +4602,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous and mostly title-similar rather than truly duplicate: the soft edges span unrelated tokenizer, multimodal, training, loading, and quantization bugs, so none should be merged.", - "confidence": 0.94, - "canonical_issue_reason": "No clear canonical issue: the items do not share one underlying defect, so a single representative would be arbitrary.", + "summary": "The candidate pairs are mostly topic-adjacent but not duplicate-level matches. They span unrelated tokenizer, multimodal, quantization, training, and config regressions, with no pair sharing the same concrete bug or code path closely enough to merge.", + "confidence": 0.83, + "canonical_issue_reason": "issue:42886 is the broadest representative of the tokenizer/offline-loading subtheme: it describes a clear user-visible failure, has decent discussion activity, and is more general than the narrower implementation-specific tokenizer reports.", "canonical_pr_reason": null, - "best_issue_reason": "No global best issue stands out because the cluster mixes distinct bug classes; none is clearly representative of the whole set.", + "best_issue_reason": "issue:42886 is the best anchor for this cluster because it is concrete yet broad, making it the most representative issue among otherwise disparate topics.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43531", "right": "issue:44589", "accept": false, - "reason": "Different failures: Qwen3-MoE sliding-window behavior vs Float8 storage/import lookup." + "reason": "Different failures in different paths: Qwen3-MoE sliding-window behavior vs a Float8 storage import/type error." }, { "left": "issue:44987", "right": "issue:45042", "accept": false, - "reason": "Unrelated bugs: loading a model from transformers>=5.1.0 vs PIL image processors incorrectly requiring torchvision." + "reason": "One is a model-loading regression for a specific repo; the other is a PIL image processor dependency regression." }, { "left": "issue:43782", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen VL issues, but one is a from_pretrained weight_only load error and the other is a video position-id regression; not the same bug." + "reason": "Both involve Qwen VL models, but the bugs differ: weight_only loading vs wrong vision position ids for video input." }, { "left": "issue:44077", "right": "issue:44625", "accept": false, - "reason": "Different code paths: patchtsmixer post_init validation vs Qwen3.5 num_labels propagation into text config." + "reason": "Unrelated model/config issues: patchtsmixer post_init handling vs Qwen3.5 num_labels propagation." }, { "left": "issue:45759", "right": "issue:45859", "accept": false, - "reason": "Config unwrapping for composite Qwen multimodal models is unrelated to CPU-offload compatibility in rotary embeddings." + "reason": "Different code paths: composite config unwrapping vs CPU-offload incompatibility in rotary embeddings." }, { "left": "issue:44918", "right": "issue:44936", "accept": false, - "reason": "One is TRL SFT input embedding unpacking, the other is trainer.evaluate() after trainer.train(); distinct trainer lifecycle bugs." - }, - { - "left": "issue:44368", - "right": "issue:45357", - "accept": false, - "reason": "Both mention Qwen3.5, but one is a LoRA tie_word_embeddings warning and the other is incorrect visual-encoder keys on save_pretrained." + "reason": "Trainer-related, but not the same bug: embedding unpacking in SFT vs evaluate() failing after train()." }, { "left": "issue:44062", "right": "issue:44451", "accept": false, - "reason": "AddedToken special-argument duplication is unrelated to the ScandiBERT load failure." + "reason": "Tokenizer-related, but one is an AddedToken argument error while the other is a model loading failure for ScandiBERT." }, { - "left": "issue:39692", - "right": "issue:44610", + "left": "issue:44368", + "right": "issue:45357", "accept": false, - "reason": "Different models and failures: SigLIP2 doc example/quantization issues vs OmDet-Turbo processor size mismatch." + "reason": "A warning about tie_word_embeddings is unrelated to incorrect visual encoder key saving." }, { "left": "issue:45588", "right": "issue:45759", "accept": false, - "reason": "Flash-attention sink-less model crash is a different bug from Qwen multimodal config unwrapping." + "reason": "Flash-attention sink handling and multimodal config unwrapping are different bugs." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Both mention model/processor mismatches, but they affect different models and different failure modes." }, { "left": "issue:45704", "right": "issue:45923", "accept": false, - "reason": "T5 apex RMSNorm memory leak is unrelated to the missing supports_gradient_checkpointing flag in a trust_remote_code model." + "reason": "A memory leak in apex RMSNorm is unrelated to a missing gradient-checkpointing flag." }, { "left": "issue:43746", "right": "issue:46050", "accept": false, - "reason": "PEFT local-checkpoint loading failure and sinq save/load corruption are distinct issues." + "reason": "Different load/save regressions in different mechanisms: PEFT local checkpoints vs sinq quantized round-trip." }, { "left": "issue:43421", "right": "issue:43653", "accept": false, - "reason": "Both involve tokenizers and special tokens, but one is runtime post-processor updates and the other is BigBirdTokenizer mask-token registration/empty decode output; not the same concrete defect." + "reason": "Both concern special tokens, but one is runtime post-processor updates and the other is BigBird mask-token registration." }, { "left": "issue:45663", "right": "issue:45923", "accept": false, - "reason": "Gemma-4 FSDP2 shared_kv_states KeyError is unrelated to the gradient-checkpointing flag missing on a trust_remote_code variant." + "reason": "Gemma-4 FSDP2 attention state reuse is unrelated to a model capability flag missing on trust_remote_code." }, { "left": "issue:42886", "right": "issue:44843", "accept": false, - "reason": "Both hit offline mode, but 44843 is specifically model_info() called in _patch_mistral_regex; 42886 is a broader tokenizer cache-load failure." + "reason": "Related offline/tokenizer theme, but not the same concrete bug: generic cache loading failure vs a specific mistral-regex patch call path." }, { "left": "issue:43381", "right": "issue:43844", "accept": false, - "reason": "Eval-mode gradient-checkpointing guard and ZeRO-3 gradient blow-up are different training issues." + "reason": "Gradient checkpointing in eval mode is a different issue from abnormal gradient growth under ZeRO-3." }, { "left": "issue:43421", "right": "issue:44186", "accept": false, - "reason": "Special-token post-processor refresh is unrelated to LayoutLMv2Tokenizer crashes on NER padding/truncation." + "reason": "Special-token post-processor behavior does not match a LayoutLMv2Tokenizer crash on NER padding/truncation." }, { "left": "issue:43531", "right": "issue:43645", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior is unrelated to custom model initialization breakage in Jupyter notebooks." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to Jupyter custom-model initialization in Transformers 5.0." }, { "left": "issue:43994", "right": "issue:44671", "accept": false, - "reason": "Different model regressions: SigLIP2 nonsensical outputs vs CamemBERT masked-LM predictions in v5." + "reason": "Both are wrong-output regressions, but they affect different models and distinct prediction paths." }, { - "left": "issue:45588", - "right": "issue:45759", + "left": "issue:45704", + "right": "issue:45859", "accept": false, - "reason": "Flash-attention crash and Qwen config unwrapping are unrelated bugs." + "reason": "A T5 RMSNorm memory leak is unrelated to CPU offload incompatibility in Qwen3_5 MoE rotary embedding." }, { "left": "issue:41720", "right": "issue:45704", "accept": false, - "reason": "cudaErrorAssert with Qwen3 auto device mapping is unrelated to the T5 RMSNorm memory leak." + "reason": "CUDA assert during Qwen3 auto device mapping is not the same bug as the T5 memory leak." }, { "left": "issue:45081", "right": "issue:45356", "accept": false, - "reason": "Same general tokenizer-regex area, but the symptoms differ: hard _patch_mistral_regex crash vs codec-handling regression with a misleading warning; not enough evidence of one underlying bug." + "reason": "Both touch fix_mistral_regex, but one crashes with AttributeError while the other reports codec-handling regression and a misleading warning." }, { "left": "issue:43596", "right": "issue:44220", "accept": false, - "reason": "ZeRO-3/BertModel index error is unrelated to _torch_extract_fbank_features()." + "reason": "BERT zero3 init IndexError and audio fbank feature extraction are unrelated." }, { "left": "issue:43746", "right": "issue:43873", "accept": false, - "reason": "PEFT checkpoint loading and quantization/offloading behavior are different problems." + "reason": "Local checkpoint loading with PEFT is not the same as quantization offloading behavior." }, { "left": "issue:39692", "right": "issue:43873", "accept": false, - "reason": "SigLIP2 doc-example failures are unrelated to offloading not working with quantization." + "reason": "A SigLIP2 doc/model mismatch and quantization failure is unrelated to offloading with quantization." }, { - "left": "issue:40444", - "right": "issue:43746", + "left": "issue:45356", + "right": "issue:45907", "accept": false, - "reason": "Qwen2.5-VL multi-image IterableDataset fine-tuning and GraniteSpeech PEFT checkpoint loading are separate bugs." + "reason": "Tokenizer codec/regression issues are unrelated to forwarding the revision argument in list_repo_files." }, { - "left": "issue:45356", - "right": "issue:45907", + "left": "issue:40444", + "right": "issue:43746", "accept": false, - "reason": "Tokenizer regression is unrelated to list_repo_files failing to forward revision." + "reason": "Multi-image iterable dataset finetuning fails for Qwen2.5-VL; PEFT local checkpoint loading is a different problem." }, { "left": "issue:44220", "right": "issue:44661", "accept": false, - "reason": "Audio feature extraction bug and add-new-model-like/tokenizer mapping issue are distinct." + "reason": "Audio feature extraction and add-new-model-like mapping logic are different subsystems and bugs." }, { "left": "issue:30064", "right": "issue:36010", "accept": false, - "reason": "Void segmentation map processing and GenerationMixin import failure have no shared code path." + "reason": "Image-processor void segmentation maps and a GenerationMixin import error have no common code path." }, { "left": "issue:43720", "right": "issue:43819", "accept": false, - "reason": "BitNet packed-weight loading and DAC.from_latents/STE mismatch are unrelated implementation bugs." + "reason": "BitNet packed-weight loading and DAC latent reconstruction/STE behavior are unrelated." } ] }, @@ -4750,8 +4806,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -4805,16 +4861,16 @@ ], "soft_pairs": [ "issue:43540|issue:44008", - "issue:43618|issue:43697", "issue:43653|issue:43720", + "issue:43618|issue:43697", "issue:45183|issue:45479", "issue:43388|issue:46032", - "issue:43716|issue:44008", "issue:43550|issue:45072", + "issue:43716|issue:44008", "issue:43295|issue:45834", "issue:44062|issue:44488", - "issue:43335|issue:43756", "issue:43550|issue:43819", + "issue:43335|issue:43756", "issue:43322|issue:45685", "issue:43950|issue:45987", "issue:44186|issue:44898", @@ -4825,203 +4881,203 @@ "issue:43540|issue:45381", "issue:44448|issue:44671", "issue:43577|issue:43818", - "issue:43782|issue:44977", - "issue:43421|issue:44811", "issue:30064|issue:36331", + "issue:43421|issue:44811", + "issue:43782|issue:44977", "issue:44821|issue:45084", "issue:41720|issue:44991", "issue:39692|issue:45834" ], - "prepared_review_unit_hash": "b4a31c34395c88d1c1bdf39ccd260562fb669abb8a76dbc8fae474449dd3c552", + "prepared_review_unit_hash": "e411b3b124260044dd9bf24b1b5e948f49c46161d9d7b1d1175cf65e03466ab9", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35665, - "estimated_input_tokens": 8917, - "estimated_eval_tokens": 18090 + "serialized_chars": 35676, + "estimated_input_tokens": 8919, + "estimated_eval_tokens": 18094 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: the issues cover unrelated bugs in tokenizers, multimodal processors, config loading, caching, training, and model-specific inference paths. None of the suggested soft pairs look like the same underlying defect, so they should not be merged as duplicates.", - "confidence": 0.98, - "canonical_issue_reason": "No canonical issue: the items span unrelated regressions across different models/subsystems, and the soft pairs are only superficially similar.", - "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good global anchor for this cluster, because there is no coherent duplicate set here.", - "best_pr_reason": null, + "summary": "No duplicate cluster emerges here: the soft pairs span unrelated tokenizer, vision, training, cache, and model-loading regressions. All proposed pairings should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits this set; the items are heterogeneous and the soft pairs do not describe the same underlying defect.", + "canonical_pr_reason": "No pull requests are present in the provided cluster.", + "best_issue_reason": "No global best issue: each issue is model- or component-specific, so none cleanly represents the others as a duplicate anchor.", + "best_pr_reason": "No pull requests are present in the provided cluster.", "soft_edge_verdicts": [ { "left": "issue:43540", "right": "issue:44008", "accept": false, - "reason": "Different models and failures: Qwen3OmniMoe video-input ValueError vs Gemma 3n forward variable collision." + "reason": "Different multimodal failure modes: Qwen3OmniMoe video processing vs a Gemma3n forward variable-name collision." }, { - "left": "issue:43618", - "right": "issue:43697", + "left": "issue:43653", + "right": "issue:43720", "accept": false, - "reason": "Both are output-related regressions, but on different models and code paths (CLIP attentions vs RTDetr v5 outputs)." + "reason": "Unrelated code paths: tokenizer special-token registration vs BitNet packed-weight loading in accelerate." }, { - "left": "issue:43653", - "right": "issue:43720", + "left": "issue:43618", + "right": "issue:43697", "accept": false, - "reason": "Tokenizer special-token registration bug vs BitNet packed-weight loading bug; unrelated mechanisms." + "reason": "Both are v5 regressions, but one is CLIP output field assignment and the other is RTDetr output divergence." }, { "left": "issue:45183", "right": "issue:45479", "accept": false, - "reason": "Validation/XOR error-message bug is not the same as the degenerate zero-loss classification bug." + "reason": "Different bug classes: validation/error-message logic vs an actual zero-loss classification bug." }, { "left": "issue:43388", "right": "issue:46032", "accept": false, - "reason": "Metric gathering label truncation and Mamba2 use_cache correctness are distinct runtime paths." + "reason": "Different runtime areas: distributed metric gathering vs Mamba2 cache correctness with seq_len > 1." }, { - "left": "issue:43716", - "right": "issue:44008", + "left": "issue:43550", + "right": "issue:45072", "accept": false, - "reason": "Dtype mismatch in Mistral-3 preprocessing is unrelated to Gemma 3n audio_mel_mask name collision." + "reason": "Separate failures: torch.compile+SDPA behavior vs dtype mismatch during bfloat16 inference." }, { - "left": "issue:43550", - "right": "issue:45072", + "left": "issue:43716", + "right": "issue:44008", "accept": false, - "reason": "Different models and symptoms: torch.compile/SDPA failure vs bfloat16 dtype mismatches." + "reason": "Different root causes: image preprocessor/model dtype mismatch vs an AttributeError from a forward variable collision." }, { "left": "issue:43295", "right": "issue:45834", "accept": false, - "reason": "Custom-model processor/tokenizer regression is not the same as Kosmos2.5 long-OCR index error." + "reason": "Different surfaces: processor/tokenizer regression vs a long-OCR index error in Kosmos2.5." }, { "left": "issue:44062", "right": "issue:44488", "accept": false, - "reason": "AddedToken special-arg duplication is unrelated to loading cjvt/sleng-bert." - }, - { - "left": "issue:43335", - "right": "issue:43756", - "accept": false, - "reason": "SwitchTransformers sparse-layer creation bug vs Smollm3 RoPE-layer dropping are different configuration issues." + "reason": "One is an AddedToken argument collision; the other is a tokenizer load failure for a specific model." }, { "left": "issue:43550", "right": "issue:43819", "accept": false, - "reason": "Both are model bugs, but Bamba SDPA/compile and DAC from_latents mismatch are separate code paths." + "reason": "Not the same bug: compile/SDPA failure vs DAC latent decoding not matching forward pass." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Different model-specific configuration mistakes: sparse-layer creation vs RoPE-layer count mismatch." }, { "left": "issue:43322", "right": "issue:45685", "accept": false, - "reason": "Llava-Next loading segfault and moe MPS histogram error are unrelated failures." + "reason": "Completely different domains: Llava Next load crash vs an MPS histogram/int type error in moe code." }, { "left": "issue:43950", "right": "issue:45987", "accept": false, - "reason": "from_pretrained buffer corruption is not the same as StaticCache.get_seq_length() returning the wrong type." + "reason": "Unrelated issues: corrupted non-persistent buffers on load vs return-type contract mismatch in StaticCache." }, { "left": "issue:44186", "right": "issue:44898", "accept": false, - "reason": "LayoutLMv2 tokenizer batching crash and Perceiver image-classification resolution bug are distinct." + "reason": "Tokenizer/padding crash vs image-classification interpolation failure; same broad area, different defects." }, { "left": "issue:44843", "right": "issue:45081", "accept": false, - "reason": "Offline model_info() call in tokenizer patching is different from backend_tokenizer AttributeError in the same patch helper." + "reason": "Same helper name appears, but one is offline model_info access and the other is a backend_tokenizer AttributeError." }, { "left": "issue:43653", "right": "issue:43819", "accept": false, - "reason": "Tokenizer special-token registration and DAC STE mismatch have no shared underlying bug." + "reason": "Tokenizer special-token handling is unrelated to DAC from_latents / STE matching." }, { "left": "issue:43531", "right": "issue:43643", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and AutoConfig trust_remote_code field loss are unrelated." + "reason": "Qwen3-MoE sliding-window behavior and AutoConfig trust_remote_code field loss are unrelated bugs." }, { "left": "issue:43643", "right": "issue:44877", "accept": false, - "reason": "Both involve config loading, but one is trust_remote_code field loss and the other is strict config rejection for granite_speech." + "reason": "Both concern config loading, but one is missing fields from trust_remote_code and the other is strict config rejection of granite_speech." }, { "left": "issue:43540", "right": "issue:45381", "accept": false, - "reason": "Both mention Qwen video inputs, but they affect different models and defects (processing error vs wrong vision_position_ids)." + "reason": "Different multimodal issues: video input processing vs wrong vision_position_ids in qwen2.5-vl." }, { "left": "issue:44448", "right": "issue:44671", "accept": false, - "reason": "Different model outputs across versions, but Pegasus generation drift and CamemBERT masked-LM regression are separate bugs." + "reason": "Both are v4/v5 output regressions, but they affect different models and different inference paths." }, { "left": "issue:43577", "right": "issue:43818", "accept": false, - "reason": "BLIP2 dtype propagation and Video-LLaVA temporal-attention/weight-sharing issues are unrelated." + "reason": "One is dtype propagation in Blip2; the other is a Video-LLaVA architecture/loading issue." }, { - "left": "issue:43782", - "right": "issue:44977", + "left": "issue:30064", + "right": "issue:36331", "accept": false, - "reason": "Qwen3VL weight_only load error is not the same as Qwen3.5 flash-attention generation failure." + "reason": "Image segmentation-map processing and CustomTrainer compute_loss argument handling are unrelated." }, { "left": "issue:43421", "right": "issue:44811", "accept": false, - "reason": "Tokenizer post-processor update feature gap is unrelated to Whisper batch_decode skip_special_tokens handling." + "reason": "Tokenizer post-processor updates and Whisper batch_decode skip_special_tokens are different tokenizer behaviors." }, { - "left": "issue:30064", - "right": "issue:36331", + "left": "issue:43782", + "right": "issue:44977", "accept": false, - "reason": "Image processor segmentation-map failure and CustomTrainer compute_loss signature mismatch are unrelated." + "reason": "Qwen3VL weight_only loading error vs flash-attention generation bug in Qwen3.5; different failure points." }, { "left": "issue:44821", "right": "issue:45084", "accept": false, - "reason": "AutoImageProcessor URL loading and template-node compilation errors have different causes and subsystems." + "reason": "AutoImageProcessor URL loading and template-node compilation errors are unrelated." }, { "left": "issue:41720", "right": "issue:44991", "accept": false, - "reason": "Qwen3 device-map cuda assert and tokenizer loading failure for est-roberta are unrelated." + "reason": "CUDA assert in Qwen3 auto device mapping is not the same defect as tokenizer loading for est-roberta." }, { "left": "issue:39692", "right": "issue:45834", "accept": false, - "reason": "SigLIP2 documentation example errors do not match Kosmos2.5 long-OCR index errors." + "reason": "SigLIP2 documentation errors and Kosmos2.5 OCR index errors are different issues." } ] }, @@ -5040,8 +5096,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -5097,8 +5153,8 @@ "issue:44561|issue:45020", "issue:45584|issue:45834", "issue:44485|issue:44509", - "issue:43644|issue:43749", "issue:43577|issue:43582", + "issue:43644|issue:43749", "issue:44186|issue:44442", "issue:44805|issue:44936", "issue:44479|issue:45834", @@ -5109,223 +5165,226 @@ "issue:44671|issue:45701", "issue:28218|issue:29159", "issue:44610|issue:45072", - "issue:43782|issue:45325", "issue:43720|issue:45072", + "issue:43782|issue:45325", "issue:42175|issue:45081", "issue:43749|issue:44291", - "issue:45081|issue:45406", "issue:43582|issue:44857", + "issue:45081|issue:45406", "issue:43650|issue:43906", "issue:43262|issue:43819", - "issue:44060|issue:44368", "issue:43526|issue:43697", - "issue:43526|issue:44554", + "issue:44060|issue:44368", "issue:34634|issue:36331", + "issue:43526|issue:44554", "issue:45084|issue:45538" ], - "prepared_review_unit_hash": "a4289899a347eaad56dd11117a9a6738bdf5e70f17fa51e82b69c7c827c99eb9", + "prepared_review_unit_hash": "4b8ab5bb968b2941bd0b6f7a88c8b2d5f854d9e4338d089bf84613ed476c8d93", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 28, - "serialized_chars": 35695, - "estimated_input_tokens": 8924, - "estimated_eval_tokens": 18104 + "serialized_chars": 35699, + "estimated_input_tokens": 8925, + "estimated_eval_tokens": 18106 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: most items are distinct Transformers bugs/regressions across tokenizers, model loading, vision/audio processors, and training. I did not find any pair that clearly represents the same underlying defect, so no soft-edge merge is justified.", - "confidence": 0.89, - "canonical_issue_reason": "If one issue must serve as a representative anchor, issue 45020 is the broadest umbrella item: it describes recent Transformers versions breaking remote_code models and is the closest to a general regression theme. That said, the cluster is not a true duplicate set.", + "summary": "Almost all pairs describe different models, subsystems, or failure modes. The only likely duplicate is the tokenizer added-token spacing report versus the slow/fast tokenizer inconsistency report.", + "confidence": 0.71, + "canonical_issue_reason": "issue:28218 is the most concrete, reproducible tokenizer bug and best captures the likely duplicate with 29159.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45020 is the best single issue candidate because it is the most broadly scoped and most likely to subsume related version-regression reports, even though the surrounding issues are not duplicates.", + "best_issue_reason": "issue:28218 is the clearest canonical target: specific symptom, easy to reproduce, and more precise than the broader inconsistency report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44561", "right": "issue:45020", "accept": false, - "reason": "Both mention v5/remote_code breakage, but one is about removal of is_torch_fx_available and the other is a broader regression report; different concrete failure modes." + "reason": "Related remote_code breakage, but 45020 is a broad umbrella report while 44561 is one specific missing-utility failure; not the same concrete bug." }, { "left": "issue:45584", "right": "issue:45834", "accept": false, - "reason": "Empty transcription after align_special_tokens is a Whisper generation bug; Kosmos2.5 long OCR input is a separate indexing issue." + "reason": "Whisper empty-transcription failure and Kosmos2.5 long-OCR index error are unrelated models and code paths." }, { "left": "issue:44485", "right": "issue:44509", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to docs still mentioning removed pipeline tasks." + "reason": "RoPE implementation discussion vs docs still mentioning removed pipeline tasks; different issues." }, { - "left": "issue:43644", - "right": "issue:43749", + "left": "issue:43577", + "right": "issue:43582", "accept": false, - "reason": "Non-persistent buffer junk in Transformers 5.0 is a state-init bug; FSDP_CPU_RAM_EFFICIENT_LOADING is a separate loading-path regression." + "reason": "Blip2 dtype loading bug and Apple Silicon allocator TypeError are unrelated." }, { - "left": "issue:43577", - "right": "issue:43582", + "left": "issue:43644", + "right": "issue:43749", "accept": false, - "reason": "Blip2 dtype staying float32 and an Apple Silicon TypeError are different initialization/device issues." + "reason": "Both concern loading, but one is non-persistent buffers in v5 and the other is FSDP CPU RAM efficient loading; different bugs." }, { "left": "issue:44186", "right": "issue:44442", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash and FastSpeech2ConformerTokenizer loading failure are unrelated tokenizer bugs." + "reason": "Different tokenizer failures on different models; no shared concrete code path." }, { "left": "issue:44805", "right": "issue:44936", "accept": false, - "reason": "An index/mask shape mismatch during model execution is not the same as trainer.evaluate() failing after training." + "reason": "Mask/index shape error and trainer.evaluate() failing after train are different failure modes." }, { "left": "issue:44479", "right": "issue:45834", "accept": false, - "reason": "Qwen video-input regression and Kosmos2.5 OCR long-input indexing are separate multimodal bugs." + "reason": "Video-input regression for Qwen VL models vs Kosmos2.5 OCR long-input error are unrelated." }, { "left": "issue:43825", "right": "issue:43994", "accept": false, - "reason": "A pipeline error message about removed translation support is not the same as siglip2 producing nonsensical outputs via AutoModel/pipeline." + "reason": "One is a wrong error message about pipeline support; the other is bad SigLIP2 outputs. Same API surface, different bug." }, { "left": "issue:43606", "right": "issue:45072", "accept": false, - "reason": "Both are dtype/device-related CI failures, but they affect different models and different mismatch mechanisms." + "reason": "Bark CPU offload device mismatch and dtype mismatches in different models are unrelated." }, { "left": "issue:43825", "right": "issue:43827", "accept": false, - "reason": "One is a runtime error-message regression; the other is documentation lag after pipeline removals." + "reason": "Both mention removed pipeline tasks, but one is runtime messaging and the other is documentation cleanup." }, { "left": "issue:34634", "right": "issue:36010", "accept": false, - "reason": "BarkProcessor voice_preset failure and GenerationMixin import failure are unrelated API breakages." + "reason": "BarkProcessor voice_preset bug and GenerationMixin import error are unrelated." }, { "left": "issue:44671", "right": "issue:45701", "accept": false, - "reason": "CamemBERT masked LM predictions and generic tokenization changes across versions are not the same bug." + "reason": "CamemBERT masked-LM regression and broad tokenization-change report are not the same concrete bug." }, { "left": "issue:28218", "right": "issue:29159", - "accept": false, - "reason": "Extra space after added token and inconsistent slow/fast tokenizer behavior overlap in area but not in concrete root cause." + "accept": true, + "reason": "Both point to tokenizer behavior diverging around added tokens/slow-vs-fast handling; the spacing bug is a concrete manifestation of the inconsistency." }, { "left": "issue:44610", "right": "issue:45072", "accept": false, - "reason": "OmDet-Turbo processor size mismatch and dtype mismatches in inference are different model/config problems." + "reason": "OmDet processor size mismatch and unrelated dtype mismatch in other models are different bugs." }, { - "left": "issue:43782", - "right": "issue:45325", + "left": "issue:43720", + "right": "issue:45072", "accept": false, - "reason": "Qwen3VL weight_only loading error and Qwen2.5-VL rope index scaling are separate loading vs position-indexing defects." + "reason": "Packed-weight loading failure and dtype mismatch inference failure are unrelated." }, { - "left": "issue:43720", - "right": "issue:45072", + "left": "issue:43782", + "right": "issue:45325", "accept": false, - "reason": "BitNet packed-weight unpacking failure and bfloat16 dtype mismatches are different accelerate-loading issues." + "reason": "Different Qwen VL bugs: one is from_pretrained weight_only loading, the other is rope/index scaling." }, { "left": "issue:42175", "right": "issue:45081", "accept": false, - "reason": "Missing tensorflow in a torch extra install and a Mistral tokenizer regex patch crash are unrelated packaging/tokenizer problems." + "reason": "Backend extras packaging issue vs Mistral regex patch crash are unrelated." }, { "left": "issue:43749", "right": "issue:44291", "accept": false, - "reason": "FSDP CPU RAM efficient loading and init_empty_weights _is_hf_initialized argument handling are separate loading mechanisms." + "reason": "FSDP CPU RAM efficient loading and init_empty_weights argument handling are different loading regressions." }, { - "left": "issue:45081", - "right": "issue:45406", + "left": "issue:43582", + "right": "issue:44857", "accept": false, - "reason": "Mistral regex patch backend-tokenizer crash and Gemma4Processor missing _tokenizer in serve are different processor/tokenizer failures." + "reason": "Apple Silicon allocator TypeError and CUDA AMP loss crash are unrelated." }, { - "left": "issue:43582", - "right": "issue:44857", + "left": "issue:45081", + "right": "issue:45406", "accept": false, - "reason": "Apple Silicon TypeError in caching_allocator_warmup and CUDA AMP float16 crash in LwDetrImageLoss are unrelated runtime issues." + "reason": "Tokenizer patch AttributeError and Gemma4Processor missing _tokenizer are different attribute-access bugs." }, { "left": "issue:43650", "right": "issue:43906", "accept": false, - "reason": "The contentless 'ADD THE DATA' issue is not a reproduction of another bug report." + "reason": "A placeholder/low-content issue and a reproduction of another issue are not the same bug." }, { "left": "issue:43262", "right": "issue:43819", "accept": false, - "reason": "Audio processor chat-template sampling-rate defaulting and DAC.from_latents missing STE mismatch are different audio-stack bugs." - }, - { - "left": "issue:44060", - "right": "issue:44368", - "accept": false, - "reason": "A false tied-weights warning for Qwen3-Next is unrelated to a user seeing tie_word_embeddings warnings during LoRA fine-tuning." + "reason": "Audio chat-template sample-rate default and DAC from_latents mismatch are unrelated." }, { "left": "issue:43526", "right": "issue:43697", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels truncation and RTDetrV2 output changes are different vision preprocessing/model-behavior bugs." + "reason": "BeitImageProcessorFast label reduction bug and RTDetrV2 output regression are different issues." }, { - "left": "issue:43526", - "right": "issue:44554", + "left": "issue:44060", + "right": "issue:44368", "accept": false, - "reason": "Beit label reduction and the MPS attention correctness issue are unrelated." + "reason": "Both involve tied-weights warnings, but one is a specific wrong tie warning and the other is a model-config warning during fine-tuning; not the same bug." }, { "left": "issue:34634", "right": "issue:36331", "accept": false, - "reason": "BarkProcessor voice_preset and CustomTrainer.compute_loss unexpected num_items_in_batch are unrelated failures." + "reason": "BarkProcessor voice_preset issue and CustomTrainer compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "Image-processor label reduction and MPS attention correctness are unrelated." }, { "left": "issue:45084", "right": "issue:45538", "accept": false, - "reason": "Cannot compile non-template nodes and CLIPTokenizer model_max_length=10**30 are unrelated tokenizer/config issues." + "reason": "Template-compilation TypeError and CLIPTokenizer model_max_length bug are unrelated." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The analyst stays grounded in the packet: the singleton likely duplicate pair (28218/29159) is a defensible tokenizer match, and the rest of the verdicts are conservatively negative given the titles/metadata. No soft-edge acceptance appears overstated from the available evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -5337,8 +5396,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -5403,15 +5462,15 @@ "issue:36010|issue:43645", "issue:43550|issue:44912", "issue:38175|issue:44610", - "issue:39692|issue:44912", "issue:43819|issue:43927", - "issue:43257|issue:43329", + "issue:39692|issue:44912", "issue:43854|issue:44977", + "issue:43257|issue:43329", "issue:43454|issue:43577", "issue:43720|issue:44610", "issue:43531|issue:45042", - "issue:43257|issue:43696", "issue:36010|issue:36331", + "issue:43257|issue:43696", "issue:43262|issue:44811", "issue:43329|issue:44479", "issue:42490|issue:43257", @@ -5423,202 +5482,202 @@ "issue:43606|issue:45859", "issue:43257|issue:43931" ], - "prepared_review_unit_hash": "27699b4f4e8d426a8d2371748b0cf4f978dd16dc2090d99cc9ca49f1a28b3315", + "prepared_review_unit_hash": "54a42e6cdd5227154f291a4b2560383dc8ec1f19e35d0a8650b091068c85d406", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36978, - "estimated_input_tokens": 9245, - "estimated_eval_tokens": 18746 + "serialized_chars": 36973, + "estimated_input_tokens": 9244, + "estimated_eval_tokens": 18744 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab bag of unrelated issue reports. The only strong near-duplicate is the video multimodal token-count regression pair; everything else looks like separate bugs in different code paths.", - "confidence": 0.69, - "canonical_issue_reason": "issue:43329 is the best anchor because it is a concrete code-path bug in multimodal/video token accounting and is the closest root-cause-style report among the near matches.", + "summary": "Mostly a heterogeneous issue set, but one likely duplicate pair stands out around the Qwen/VL video-input regression; the rest are unrelated bugs across different models and subsystems.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 44479 is the best anchor: it describes a broad, user-facing v5.3.0 video-input regression across several Qwen VL variants and matches the shared multimodal video-path bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44479 is the best user-facing representative because it captures the broader regression impact across multiple VLM families, while still pointing at the same video-input path as the root-cause bug.", + "best_issue_reason": "44479 is the broadest and most representative issue in the only plausible duplicate thread; it captures the common regression impact more clearly than the narrower root-cause report.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43653", "right": "issue:44912", "accept": false, - "reason": "Different subsystems: tokenizer special-token registration vs quantization fallback/loading behavior." + "reason": "Different problems: BigBirdTokenizer special-token registration vs GPT-oss MXFP4 quantization fallback." }, { "left": "issue:43742", "right": "issue:45020", "accept": false, - "reason": "A specific model-load key error is not the same as the broader remote_code regression." + "reason": "Both involve loading, but one is a specific key error for MobileLLM-125M; the other is a broad remote_code regression." }, { "left": "issue:43329", "right": "issue:45600", "accept": false, - "reason": "Video token-count bug vs an auto-mapping reference to removed configs; unrelated code paths." + "reason": "Different code paths: multimodal video token counting vs stale auto_mappings references to removed configs." }, { "left": "issue:43638", "right": "issue:44936", "accept": false, - "reason": "Different training failures: ZeRO-3/non-pretrained Bert index error vs evaluate-after-train behavior." + "reason": "Unrelated training failures: deepspeed zero3/Bert indexing vs trainer.evaluate() after train()." }, { "left": "issue:45459", "right": "issue:45803", "accept": false, - "reason": "Both involve exception handling, but they are different tokenizer code paths and bugs." + "reason": "Both are exception-handling bugs, but in different functions and with different failure modes; not the same change." }, { "left": "issue:44589", "right": "issue:45042", "accept": false, - "reason": "Unrelated failures: float8 storage lookup vs PIL backend unexpectedly requiring torchvision." + "reason": "Different subsystems: storage lookup for float8 vs image processor torchvision dependency handling." }, { "left": "issue:43452", "right": "issue:44991", "accept": false, - "reason": "Both are loading-related, but one is gguf_file handling and the other is a tokenizer regression for a specific model." + "reason": "Both are from_pretrained loading issues, but they affect different mechanisms and models (gguf_file vs tokenizer regression)." }, { "left": "issue:43746", "right": "issue:44811", "accept": false, - "reason": "Different areas: local checkpoint loading with PEFT adapters vs Whisper batch_decode skip_special_tokens handling." + "reason": "Local checkpoint loading with PEFT adapters is unrelated to Whisper batch_decode skipping special tokens." }, { "left": "issue:36010", "right": "issue:43645", "accept": false, - "reason": "ImportError for GenerationMixin is unrelated to the Jupyter custom-model initialization regression." + "reason": "ImportError for GenerationMixin is not the same as Jupyter/custom-model initialization breakage in Transformers 5.0." }, { "left": "issue:43550", "right": "issue:44912", "accept": false, - "reason": "Different concrete bugs: torch.compile/SDPA failure vs MXFP4 quantization fallback." + "reason": "Different bugs: torch.compile/SDPA failure in Bamba vs quantization fallback in GPT-oss." }, { "left": "issue:38175", "right": "issue:44610", "accept": false, - "reason": "Different models and symptoms: zero probabilities in SigLIP2 vs processor output size mismatch in OmDet-Turbo." - }, - { - "left": "issue:39692", - "right": "issue:44912", - "accept": false, - "reason": "Docs/example issues for SigLIP2 are not the same as GPT-oss quantization loading behavior." + "reason": "Different model/processor issues: zero probabilities in SigLIP2 vs wrong image size in OmDet-Turbo." }, { "left": "issue:43819", "right": "issue:43927", "accept": false, - "reason": "Different bugs: DAC latent reconstruction mismatch vs config save/load losing token IDs." + "reason": "DAC forward/from_latents mismatch is unrelated to DiaConfig save/load custom token IDs." }, { - "left": "issue:43257", - "right": "issue:43329", + "left": "issue:39692", + "right": "issue:44912", "accept": false, - "reason": "MoE weight conversion during accelerate/deepspeed loading is unrelated to video token counting." + "reason": "SigLIP2 doc example errors are unrelated to GPT-oss quantization fallback." }, { "left": "issue:43854", "right": "issue:44977", "accept": false, - "reason": "Different models and failure modes: GLM-4.7-Flash load issue vs Qwen3.5 flash-attention generation bug." + "reason": "Different model-specific generation/loading issues: GLM-4.7-Flash test loading vs Qwen3.5 flash-attention generation." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE accelerate/deepspeed weight conversion is unrelated to multimodal video token counting." }, { "left": "issue:43454", "right": "issue:43577", "accept": false, - "reason": "Weight-tying garbage generation and dtype propagation are separate loading/model-init bugs." + "reason": "AyaVision lm_head tying and Blip2 dtype propagation are distinct model-loading defects." }, { "left": "issue:43720", "right": "issue:44610", "accept": false, - "reason": "Packed-weight unpacking during loading is unrelated to image input resolution mismatch." + "reason": "Packed-weight unpacking during accelerate loading is unrelated to OmDet-Turbo processor image size." }, { "left": "issue:43531", "right": "issue:45042", "accept": false, - "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to PIL backend/torchvision requirements." + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to PIL backend torchvision requirements." }, { - "left": "issue:43257", - "right": "issue:43696", + "left": "issue:36010", + "right": "issue:36331", "accept": false, - "reason": "MoE weight conversion vs GPU OOM are different issues." + "reason": "Importing GenerationMixin and CustomTrainer.compute_loss signature mismatch are different bugs." }, { - "left": "issue:36010", - "right": "issue:36331", + "left": "issue:43257", + "right": "issue:43696", "accept": false, - "reason": "Import path breakage is unrelated to compute_loss signature incompatibility." + "reason": "Qwen3 MoE weight conversion/loading is unrelated to GPT-oss CUDA OOM." }, { "left": "issue:43262", "right": "issue:44811", "accept": false, - "reason": "Audio chat-template sampling rate defaulting is unrelated to Whisper decode special-token handling." + "reason": "Audio chat-template sample-rate defaults and Whisper batch_decode token skipping are different processor bugs." }, { "left": "issue:43329", "right": "issue:44479", "accept": true, - "reason": "Both point to the same video-input multimodal path: a video branch/token-count regression affecting multiple VLM processors." + "reason": "Both point to the same multimodal video-input regression; 43329 names the likely root cause in the video branch, while 44479 reports the broad Qwen VL symptom set." }, { "left": "issue:42490", "right": "issue:43257", "accept": false, - "reason": "Trailing-slash local path warning is unrelated to MoE weight conversion/loading." + "reason": "Local-path slash import warning is unrelated to Qwen3 MoE weight conversion/loading." }, { "left": "issue:43749", "right": "issue:44756", "accept": false, - "reason": "Both concern loading/memory, but FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different fixes." + "reason": "FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different memory/loading problems." }, { "left": "issue:43643", "right": "issue:45759", "accept": false, - "reason": "Config loading with trust_remote_code missing fields is not the same as unwrapping text_config for composite multimodal configs." + "reason": "Both involve config loading, but one is missing fields with trust_remote_code and the other is unwrapping text_config for composite multimodal configs." }, { "left": "issue:44568", "right": "issue:45356", "accept": false, - "reason": "Different tokenizer regressions for different models and different mechanisms." + "reason": "Different tokenizer regressions affecting different models and error modes." }, { "left": "issue:43329", "right": "issue:43450", "accept": false, - "reason": "Video token accounting bug is not the same as batched video processor shape output." + "reason": "The video-token counting bug is not the same as batched video processor shape handling." }, { "left": "issue:34567", "right": "issue:40990", "accept": false, - "reason": "TrainerState token counter bug is unrelated to GPT-oss perplexity behavior." + "reason": "Trainer token-count tracking and GPT-oss perplexity are unrelated." }, { "left": "issue:43606", @@ -5630,13 +5689,13 @@ "left": "issue:43257", "right": "issue:43931", "accept": false, - "reason": "Weight conversion/loading with accelerate+deepspeed is unrelated to Qwen3-VL shape mismatch." + "reason": "Weight conversion during accelerate/deepspeed loading is unrelated to Qwen3-VL weight shape mismatch." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded overall. The canonical/best-issue choices are supported by the titles, and the only accepted soft pair (issue:43329 vs issue:44479) looks like the same video-token accounting bug at different specificity levels. The remaining soft-edge rejections are conservative and do not overclaim duplication from the available evidence." + "feedback": "The summary is grounded in the packet and stays conservative: it identifies one plausible duplicate thread around the Qwen/VL video-input regression while treating the rest as heterogeneous, which matches the titles and soft-similarity evidence. The lone accepted pair (43329 \u2194 44479) is a reasonable same-bugfix match, and the rejected pairs are framed narrowly enough to avoid overclaiming." }, "evaluator_used": true, "retried": false, @@ -5652,8 +5711,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -5714,16 +5773,16 @@ "issue:42915|issue:44821", "issue:43452|issue:43582", "issue:42915|issue:43742", - "issue:43697|issue:45357", "issue:44987|issue:45538", + "issue:43697|issue:45357", "issue:42915|issue:45538", "issue:41720|issue:43540", "issue:33357|issue:34567", "issue:44479|issue:45381", "issue:43927|issue:44912", "issue:42175|issue:43824", - "issue:44987|issue:45907", "issue:43317|issue:45538", + "issue:44987|issue:45907", "issue:43540|issue:44479", "issue:43540|issue:44560", "issue:43257|issue:43756", @@ -5732,27 +5791,27 @@ "issue:42175|issue:44938", "issue:42175|issue:44488", "issue:45137|issue:45405", - "issue:43742|issue:44291", "issue:24643|issue:29159", + "issue:43742|issue:44291", "issue:28282|issue:34689", "issue:44737|issue:45588", "issue:43404|issue:45987", "issue:38175|issue:44898" ], - "prepared_review_unit_hash": "244b99c4985ce030233cd648965b5ad6d22206b719acac80fd491856d0c7f60e", + "prepared_review_unit_hash": "b2cd293765045c8cd87d83deb9ba62965c9a66699afab61013d219fc61ae83df", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 37044, + "serialized_chars": 37041, "estimated_input_tokens": 9261, "estimated_eval_tokens": 18778 }, @@ -5761,205 +5820,202 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This is a loose, heterogeneous cluster of Transformers issues. The soft pairs mostly share broad themes like tokenization, model loading, or video/model regressions, but they do not describe the same concrete bug or fix path.", - "confidence": 0.89, - "canonical_issue_reason": "issue:45488 is the strongest representative of the only somewhat coherent subtheme here: a concrete tokenizer regression with clear user impact and inbound references. It is more actionable than the broader or more model-specific reports.", + "summary": "Mostly a set of unrelated Transformers issues; only one soft pair looks like a plausible duplicate cluster around Qwen video-input regression. All other pairs are too model-, feature-, or subsystem-specific to merge safely.", + "confidence": 0.71, + "canonical_issue_reason": "Issue 44479 is the best anchor for the only likely duplicate mini-cluster: it describes the broader Qwen video-input regression across multiple Qwen VL/MoE variants, while 45381 is a narrower symptom on qwen2.5-vl. The rest of the set does not form a coherent duplicate cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45488 is the best standalone issue in this mixed set because it is specific, reproducible, and has the clearest cross-model regression signal among the tokenization-related reports.", + "best_issue_reason": "Issue 44479 is the strongest representative issue because it is the broadest and clearest statement of the shared video-input regression, and it subsumes the narrower qwen2.5-vl symptom in 45381.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44568", "right": "issue:45488", "accept": false, - "reason": "Both involve tokenizer behavior, but one is about add_special_tokens/BOS-EOS handling while the other is a LlamaTokenizer pre-tokenizer override regression. Different underlying bugs." + "reason": "Both involve tokenizer behavior in v5, but the concrete bugs differ: missing BOS/EOS insertion vs overriding a tokenizer.json pre-tokenizer. Not the same code-path problem." }, { "left": "issue:44849", "right": "issue:45701", "accept": false, - "reason": "Both mention tokenization/version changes, but 44849 is a model-specific hidden-state bug and 45701 is a broad tokenization complaint. Not the same code path." + "reason": "One is a Qwen3.5 hidden-states bug, the other a broad tokenization-version complaint. Too vague and not the same fix." }, { "left": "issue:29159", "right": "issue:34567", "accept": false, - "reason": "Tokenizer inconsistency versus TrainerState token-count tracking. Completely different subsystems and failure modes." + "reason": "Tokenizer fast/slow inconsistency and TrainerState token counter updating are unrelated issues." }, { "left": "issue:44291", "right": "issue:45042", "accept": false, - "reason": "Loading with init_empty_weights and PIL backend requiring torchvision are unrelated problems in different parts of the stack." + "reason": "Different subsystems and failure modes: init_empty_weights argument handling vs PIL image processor dependency requirements." }, { "left": "issue:42915", "right": "issue:44821", "accept": false, - "reason": "Qwen3Moe FP8 loading failure and AutoImageProcessor URL loading are different model/loading paths." + "reason": "Qwen3Moe FP8 loading and AutoImageProcessor URL loading are unrelated." }, { "left": "issue:43452", "right": "issue:43582", "accept": false, - "reason": "gguf_file handling for tokenizer/model loading does not match the Apple Silicon warmup TypeError in caching_allocator_warmup." + "reason": "GGUF loading failure and Apple Silicon caching_allocator_warmup TypeError are unrelated code paths." }, { "left": "issue:42915", "right": "issue:43742", "accept": false, - "reason": "Both are load-time failures, but one is Qwen3Moe FP8-related and the other is a MobileLLM key error. Different bugs." + "reason": "Different model-loading failures with no shared concrete mechanism." }, { - "left": "issue:43697", - "right": "issue:45357", + "left": "issue:44987", + "right": "issue:45538", "accept": false, - "reason": "Different models and symptoms: RTDetrV2 output drift versus Qwen3.5 save_pretrained visual-key regression." + "reason": "A model-loading regression and a CLIPTokenizer max-length bug are not the same underlying issue." }, { - "left": "issue:44987", - "right": "issue:45538", + "left": "issue:43697", + "right": "issue:45357", "accept": false, - "reason": "A model loading regression for physical-intelligence/fast is not the same as CLIPTokenizer's model_max_length issue." + "reason": "Different problems: RTDetrV2 output drift vs incorrect visual encoder keys on save_pretrained." }, { "left": "issue:42915", "right": "issue:45538", "accept": false, - "reason": "Qwen3Moe FP8 loading and CLIPTokenizer max-length behavior are unrelated." + "reason": "FP8/Qwen3Moe loading and CLIP tokenizer max-length are unrelated." }, { "left": "issue:41720", "right": "issue:43540", "accept": false, - "reason": "Auto device mapping cudaErrorAssert on A800 and a Qwen3OmniMoe video-processing ValueError are different failures." + "reason": "Auto device-mapping CUDA assert and Qwen3OmniMoe video processing ValueError are different failures." }, { "left": "issue:33357", "right": "issue:34567", "accept": false, - "reason": "MacOS bus error in CLIP loading versus TrainerState token accounting. No shared bug." + "reason": "Mac bus error on CLIP loading and TrainerState token counting are unrelated." }, { "left": "issue:44479", "right": "issue:45381", - "accept": false, - "reason": "Both are video-input regressions, but they target different model families and likely different code paths." + "accept": true, + "reason": "Both point to the same Qwen video-input regression: 44479 is the broader v5.3.0 regression across Qwen VL/MoE variants, and 45381 is a narrower qwen2.5-vl manifestation involving vision_position_ids." }, { "left": "issue:43927", "right": "issue:44912", "accept": false, - "reason": "DiaConfig save/load custom token IDs and MXFP4 quantization fallback are unrelated." + "reason": "DiaConfig save/load token-ID loss and MXFP4 quantization fallback are unrelated." }, { "left": "issue:42175", "right": "issue:43824", "accept": false, - "reason": "TensorFlow backend packaging and missing Qwen2_5_VL import are different dependency/import issues." + "reason": "Backend install/dependency packaging issue vs missing Qwen2_5_VL symbol import are different bugs." }, { - "left": "issue:44987", - "right": "issue:45907", + "left": "issue:43317", + "right": "issue:45538", "accept": false, - "reason": "Loading a specific model and forwarding revision in list_repo_files are unrelated CI/API issues." + "reason": "device_map/offload dequantization failure and CLIPTokenizer model_max_length are unrelated." }, { - "left": "issue:43317", - "right": "issue:45538", + "left": "issue:44987", + "right": "issue:45907", "accept": false, - "reason": "Dequantized model offload failure and CLIPTokenizer max length are different areas of the codebase." + "reason": "Model-loading failure and list_repo_files revision forwarding are unrelated." }, { "left": "issue:43540", "right": "issue:44479", "accept": false, - "reason": "Both mention video inputs, but 43540 is Qwen3OmniMoe-specific while 44479 is a broader v5.3.0 regression across multiple model families." + "reason": "Both mention video inputs, but the model families and symptoms differ enough that they do not clearly share the same concrete bug." }, { "left": "issue:43540", "right": "issue:44560", "accept": false, - "reason": "Qwen3OmniMoe video ValueError and Qwen3-vl-embedding StopIteration are different bugs in different model implementations." + "reason": "Different Qwen video-input failures with different models and error types." }, { "left": "issue:43257", "right": "issue:43756", "accept": false, - "reason": "Qwen3 MOE weight conversion with accelerate+deepspeed is unrelated to Smollm3 RoPE-layer dropping." + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed and Smollm3 RoPE-layer dropping are unrelated." }, { "left": "issue:43577", "right": "issue:45072", "accept": false, - "reason": "Both concern dtype, but on different models and different mechanisms; not the same fixable bug." + "reason": "Both are dtype-related, but they affect different model classes and likely different fixes." }, { "left": "issue:44062", "right": "issue:44991", "accept": false, - "reason": "AddedToken special-argument duplication and tokenizer loading for EMBEDDIA/est-roberta are distinct tokenizer issues." + "reason": "AddedToken duplicate-argument error and tokenizer load failure for EMBEDDIA/est-roberta are not the same bug." }, { "left": "issue:42175", "right": "issue:44938", "accept": false, - "reason": "TensorFlow backend packaging and Python 3.14 load failure are unrelated environment issues." + "reason": "Package/backend install issue and Python 3.14 load failure are unrelated." }, { "left": "issue:42175", "right": "issue:44488", "accept": false, - "reason": "Missing TensorFlow in a torch install and failure to load cjvt/sleng-bert are different packaging/loading problems." + "reason": "TensorFlow backend packaging and loading cjvt/sleng-bert are unrelated." }, { "left": "issue:45137", "right": "issue:45405", "accept": false, - "reason": "DeepSpeed ZeRO3 deque pop error is unrelated to the PEFT minimum-version bump." + "reason": "DeepSpeed ZeRO3 deque error and PEFT minimum-version bump are unrelated." }, { "left": "issue:43742", "right": "issue:44291", "accept": false, - "reason": "MobileLLM key error and unexpected _is_hf_initialized argument during init_empty_weights are different load failures." - }, - { - "left": "issue:24643", - "right": "issue:29159", - "accept": false, - "reason": "DeepSpeed training weight-shape error versus slow/fast tokenizer inconsistency are unrelated." + "reason": "MobileLLM key error and init_empty_weights argument handling are different model-loading problems." }, { "left": "issue:28282", "right": "issue:34689", "accept": false, - "reason": "Missing PyTorch import dependency and Llama 3.2 Vision loading regression are not the same underlying issue." + "reason": "Generic PyTorch ImportError and Llama 3.2 vision model loading breakage are not the same issue." }, { "left": "issue:44737", "right": "issue:45588", "accept": false, - "reason": "XLNet CPU placement bug and flash_attention s_aux=None crash are unrelated implementation bugs." + "reason": "CPU placement bug in XLNet relative positions and flash_attention s_aux=None crash are unrelated." }, { "left": "issue:43404", "right": "issue:45987", "accept": false, - "reason": "lm_head weight tying in Mistral3 and StaticCache.get_seq_length returning the wrong shape are unrelated code paths." + "reason": "Mistral3 weight tying and StaticCache seq-length return type are unrelated code-path issues." }, { "left": "issue:38175", "right": "issue:44898", "accept": false, - "reason": "Unexpected zero probabilities in SigLIP2 and Perceiver non-default resolution failure are both vision-model bugs, but not the same defect." + "reason": "Unexpected zero probabilities in SIGLIP2 and Perceiver non-default-resolution failure are different bugs." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative. The only accepted soft pair, issue:44479 vs issue:45381, is plausibly the same Qwen video-input regression with one being a narrower manifestation of the other. The remaining pairs are reasonably rejected as different subsystems, model families, or failure modes, without overstating the evidence." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -5971,8 +6027,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -6053,196 +6109,196 @@ "issue:44514|issue:45381", "issue:44625|issue:44877" ], - "prepared_review_unit_hash": "99d450c99fec47332837f08265190bfc18f87c05c0e7be61105d32d82e75f319", + "prepared_review_unit_hash": "bf4ffe7573f2f719e4bd1a9e156f88825d7616071d81c09c9d38a2d604ac1383", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 26, - "serialized_chars": 35816, - "estimated_input_tokens": 8954, - "estimated_eval_tokens": 18164 + "serialized_chars": 35812, + "estimated_input_tokens": 8953, + "estimated_eval_tokens": 18162 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is mostly a batch of unrelated issue reports with a few superficial topic overlaps. I only found one plausible duplicate pair around SigLIP2; the rest are different models, different subsystems, or different failure modes.", - "confidence": 0.84, - "canonical_issue_reason": "issue:43493 is the clearest root-cause report in the only plausible duplicate pair: it describes the SigLIP2 implementation mismatch directly, while issue:43994 reads more like a symptom report.", + "summary": "Most pairs are not duplicates: they share broad subsystems, but not the same concrete bug. One pair appears to be the same underlying SigLIP2 model-implementation regression.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43493 is the clearest root-cause report: it identifies a concrete SigLIP2 implementation discrepancy against the original JAX model, which can subsume the symptom report in issue:43994.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43493 is the best canonical issue candidate because it most directly names the underlying bug and is more specific than the paired user-facing symptom report.", + "best_issue_reason": "issue:43493 is the best issue to anchor the cluster because it is specific, technically diagnostic, and points to the underlying defect rather than a downstream symptom.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43906", "right": "issue:45381", "accept": false, - "reason": "Different bugs and different models: a reproduction of issue 38071 vs a Qwen2.5-VL video position-id bug." + "reason": "They are both around Qwen2.5-VL-related behavior, but one is a reproduction of an external issue and the other is a video-position-id bug; no clear same code path." }, { "left": "issue:42915", "right": "issue:43749", "accept": false, - "reason": "Both involve loading/runtime problems, but one is Qwen3Moe FP8 and the other is FSDP CPU RAM-efficient loading; no shared concrete bug." + "reason": "Different failures in different loading/training paths: FineGrainedFP8Config vs FSDP CPU RAM efficient loading." }, { "left": "issue:43582", "right": "issue:43782", "accept": false, - "reason": "Unrelated failures: Apple Silicon allocator TypeError vs Qwen3VL weight_only loading error." + "reason": "AppleSilicon allocator warmup TypeError and Qwen3VL weight_only loading error are unrelated." }, { "left": "issue:28282", "right": "issue:29127", "accept": false, - "reason": "Different error classes and code paths: missing PyTorch backend vs LayoutLMv3 box-content validation messaging." + "reason": "One is a missing-PyTorch ImportError, the other is a LayoutLMv3 boxes/clarity issue; unrelated bugs." }, { "left": "issue:43526", "right": "issue:45479", "accept": false, - "reason": "One is a BEiT image-processor label reduction bug; the other is a sequence-classification loss degeneracy across models." + "reason": "BeitImageProcessorFast label reduction and sequence-classification zero-loss behavior are different code paths and symptoms." }, { "left": "issue:43540", "right": "issue:45823", "accept": false, - "reason": "Different models and symptoms: Qwen3OmniMoe video processing vs Gemma4 PLE device placement." + "reason": "Qwen3OmniMoe video-processing error and Gemma4 PLE device mismatch are different model/runtime bugs." }, { "left": "issue:44291", "right": "issue:44821", "accept": false, - "reason": "Different layers of the stack: init_empty_weights/_is_hf_initialized vs loading AutoImageProcessor from URL." + "reason": "`init_empty_weights` argument handling and loading AutoImageProcessor from URL are unrelated." }, { "left": "issue:43756", "right": "issue:45325", "accept": false, - "reason": "Both mention RoPE-related behavior, but one is Smollm3 layer dropping and the other is Qwen2.5-VL temporal position scaling; not the same bug." + "reason": "Smollm3 RoPE-layer drop and Qwen2.5-VL temporal position-id scaling are different defects." }, { "left": "issue:34567", "right": "issue:36331", "accept": false, - "reason": "TrainerState token-count tracking vs CustomTrainer compute_loss signature mismatch are separate trainer issues." + "reason": "TrainerState token-count tracking and custom compute_loss kwarg breakage are separate Trainer issues." }, { "left": "issue:43493", "right": "issue:43994", "accept": true, - "reason": "Same underlying SigLIP2 problem is plausible: both report incorrect/nonsensical outputs for the same model family, and the JAX-vs-HF discrepancy aligns with the user-facing bad-results report." + "reason": "Both describe the same SigLIP2 correctness problem: HF outputs diverge from the intended implementation and produce nonsensical results." }, { "left": "issue:44913", "right": "issue:46121", "accept": false, - "reason": "Both involve rope-related config handling, but one is a config reload defaulting bug and the other is a TypeError on JSON-loaded list validation." + "reason": "One is GPTNeoXConfig persistence of rotary_pct, the other is rope-validation input typing; related area, but not the same bug." }, { "left": "issue:44062", "right": "issue:45081", "accept": false, - "reason": "Completely different tokenizer failures: AddedToken kwargs collision vs Mistral regex patch crash." + "reason": "Different tokenizer-loading failures with different error surfaces and code paths." }, { "left": "issue:44792", "right": "issue:44964", "accept": false, - "reason": "Both are multimodal/model-loading related, but they target different models and failure modes; no concrete shared bug." + "reason": "Janus image-generation test failure and Phi-4 multimodal model loading failure are separate model-specific regressions." }, { "left": "issue:45357", "right": "issue:45920", "accept": false, - "reason": "Qwen3.5 save_pretrained visual-key regression is unrelated to the broad AutoTokenizer token-ID regression across many models." + "reason": "Qwen3.5 save_pretrained visual-key serialization and AutoTokenizer wrong token IDs are different bugs." }, { "left": "issue:43452", "right": "issue:44488", "accept": false, - "reason": "Both concern loading models/tokenizers, but one is a gguf_file regression and the other is a specific model load failure; insufficient evidence of the same bug." + "reason": "Both involve model loading, but one is gguf_file handling for AutoTokenizer/AutoModelForCausalLM and the other is a specific model load failure; not enough evidence of the same root cause." }, { "left": "issue:45042", "right": "issue:45081", "accept": false, - "reason": "Image-processor torchvision dependency regression is unrelated to a Mistral tokenizer regex crash." + "reason": "Image-processor torchvision dependency regression and Mistral regex patch tokenizer crash are unrelated." }, { "left": "issue:43742", "right": "issue:44589", "accept": false, - "reason": "KeyError while loading MobileLLM and missing Float8 storage are different loading/runtime issues." + "reason": "MobileLLM key error and missing Float8 storage object are distinct loading/runtime issues." }, { "left": "issue:43531", "right": "issue:44821", "accept": false, - "reason": "Qwen3-MoE sliding window behavior and AutoImageProcessor URL loading are unrelated." + "reason": "Qwen3-MoE sliding-window behavior and AutoImageProcessor URL loading are unrelated." }, { "left": "issue:43643", "right": "issue:45375", "accept": false, - "reason": "Both mention config fields being dropped, but one is a generic trust_remote_code AutoConfig issue and the other is a specific Qwen3.5 vision config field; not clearly the same bug." + "reason": "Both are config-field-loss complaints, but they point to different mechanisms and model-specific config schemas; not clearly the same bug." }, { "left": "issue:45405", "right": "issue:45600", "accept": false, - "reason": "A dependency version bump problem and a stale auto_mappings.py CI failure are unrelated maintenance issues." + "reason": "A version pin to unreleased PEFT and a broken auto_mappings reference are unrelated maintenance issues." }, { "left": "issue:43697", "right": "issue:45216", "accept": false, - "reason": "Different regressions: RTDetrV2 output mismatch vs Qwen3.5 save_pretrained checkpoint corruption." + "reason": "RTDetrV2 output regression and Qwen3.5 save_pretrained regression are different model bugs." }, { "left": "issue:41762", "right": "issue:44936", "accept": false, - "reason": "DeepSpeed ZeRO-3 load-time IndexError and trainer.evaluate() after train() are different execution paths." + "reason": "DeepSpeed ZeRO-3 loading crash and trainer.evaluate() after train() failure are separate Trainer/model-loading issues." }, { "left": "issue:43688", "right": "issue:43749", "accept": false, - "reason": "Auxiliary-loss normalization and FSDP CPU RAM-efficient loading are unrelated." + "reason": "Auxiliary-loss normalization and FSDP CPU RAM efficient loading are unrelated." }, { "left": "issue:43720", "right": "issue:44898", "accept": false, - "reason": "BitNet packed-weight unpacking during accelerate loading is unrelated to Perceiver interpolation behavior." + "reason": "BitNet packed-weight loading and Perceiver interpolation behavior are different code paths." }, { "left": "issue:44514", "right": "issue:45381", "accept": false, - "reason": "Both are Qwen2.5-VL, but one is batched chat-template padding and the other is video vision_position_ids; not the same concrete defect." + "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding and the other is video position-id computation; not the same bug." }, { "left": "issue:44625", "right": "issue:44877", "accept": false, - "reason": "Qwen3.5 num_labels propagation and strict config blocking granite_speech are separate config issues." + "reason": "Qwen3.5 num_labels propagation and granite_speech config strictness are different config bugs." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet, and the one accepted soft-edge pair is stated cautiously enough to fit the evidence available. The remaining negative verdicts are conservative and do not overclaim duplicate status from superficial topic overlap." + "feedback": "The cluster choice is grounded in the packet: issue:43493 is the most root-cause-like SigLIP2 report, and the accepted link to issue:43994 is a plausible same-bug pairing based on the titles. The remaining soft-edge rejections are conservative and do not overclaim duplicate status from broad subsystem similarity alone." }, "evaluator_used": true, "retried": false, @@ -6258,8 +6314,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -6320,8 +6376,8 @@ "issue:43295|issue:45356", "issue:44484|issue:44704", "issue:43819|issue:44811", - "issue:44466|issue:45005", "issue:44936|issue:45823", + "issue:44466|issue:45005", "issue:30064|issue:33666", "issue:44464|issue:44743", "issue:44361|issue:44442", @@ -6332,8 +6388,8 @@ "issue:28282|issue:33357", "issue:43653|issue:44568", "issue:43688|issue:45440", - "issue:45491|issue:46002", "issue:43577|issue:43697", + "issue:45491|issue:46002", "issue:43688|issue:45704", "issue:28282|issue:35141", "issue:44843|issue:44991", @@ -6342,14 +6398,14 @@ "issue:43646|issue:43749", "issue:29159|issue:33357" ], - "prepared_review_unit_hash": "6fc34fb0f1437537b2ef563a88156acff208379ec5629abdb4f89af84a50a9ce", + "prepared_review_unit_hash": "80a4485525d787afd03c01c2afe45959abb792a70f4609ee908ef6315e3a3c28", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, @@ -6364,192 +6420,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This looks like an accidental bundle of unrelated Transformers issues rather than a true duplicate cluster. The soft pairs mostly share broad keywords (tokenizer, v5 regressions, MoE, NaNs) but not the same concrete bug or code path.", - "confidence": 0.83, - "canonical_issue_reason": "No single canonical issue stands out because the cluster spans many unrelated regressions. If one must be chosen as an anchor, issue 43646 is the broadest v5/custom-model breakage report, but it does not actually subsume the rest.", + "summary": "This cluster is largely a grab-bag of unrelated bug reports spanning tokenizers, processors, training, serialization, and model-specific regressions. I don\u2019t see any pair that clearly meets the bar for being the same underlying bug/change.", + "confidence": 0.88, + "canonical_issue_reason": "No clear canonical issue: the items are heterogeneous, and no single report cleanly represents a true duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43646 is the closest thing to a cluster anchor because it is the most general regression report, but it is still too broad to represent the diverse failures here.", + "best_issue_reason": "Issue 44466 is the most concrete and actionable report in the set, with a specific v5 tied-weight serialization regression; if one issue must represent the cluster, this is the closest fit.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:24643", "right": "issue:29127", "accept": false, - "reason": "Different bugs: a DeepSpeed training weight-shape runtime error versus a LayoutLMv3 box-validation error." + "reason": "Different failures in different subsystems: DeepSpeed training weight-shape error vs LayoutLMv3 box-content error." }, { "left": "issue:43452", "right": "issue:43577", "accept": false, - "reason": "Both involve loading, but one is a gguf/from_pretrained breakage and the other is a dtype propagation issue in BLIP2." + "reason": "Both involve model loading, but one is a gguf/AutoTokenizer breakage and the other is an fp16/bfloat16 dtype bug in BLIP-2; not the same code-path." }, { "left": "issue:43452", "right": "issue:44843", "accept": false, - "reason": "Both touch model loading, but one is about gguf_file handling and the other is an offline model_info call in a tokenizer regex patch." + "reason": "Both touch from_pretrained, but one is gguf_file handling and the other is offline Mistral regex/model_info behavior; distinct bugs." }, { "left": "issue:44514", "right": "issue:45290", "accept": false, - "reason": "Both are chat-template crashes, but they hit different inputs and different edge cases." + "reason": "Both are chat-template related, but one is batched padding behavior and the other is tool-call assistant-message handling; too different to merge." }, { "left": "issue:44493", "right": "issue:46050", "accept": false, - "reason": "One is an unexpected position-id key serialization issue; the other is quantized save/load breakage." + "reason": "Unexpected position-id keys vs quantized save/load corruption are unrelated symptoms and code paths." }, { "left": "issue:43295", "right": "issue:45356", "accept": false, - "reason": "Both are tokenizer-related regressions, but they affect different models and different code paths." + "reason": "Both are regression-style processor/tokenizer reports, but they describe different APIs and failures, not one concrete bug." }, { "left": "issue:44484", "right": "issue:44704", "accept": false, - "reason": "A save_pretrained shard-size question is unrelated to AutoProcessor failing to forward kwargs to cached_file." + "reason": "Shard-size default behavior and AutoProcessor kwarg forwarding are unrelated issues." }, { "left": "issue:43819", "right": "issue:44811", "accept": false, - "reason": "Different subsystems: DAC latent reconstruction versus Whisper batch decoding behavior." + "reason": "DAC from_latents parity bug vs Whisper batch_decode skip_special_tokens handling are different model components and problems." }, { - "left": "issue:44466", - "right": "issue:45005", + "left": "issue:44936", + "right": "issue:45823", "accept": false, - "reason": "Both concern tied weights, but one is device-dependent lm_head serialization and the other is a broader translation-model tied-weights report." + "reason": "Trainer evaluate-after-train failure and Gemma4 device-map mismatch are separate runtime errors with different causes." }, { - "left": "issue:44936", - "right": "issue:45823", + "left": "issue:44466", + "right": "issue:45005", "accept": false, - "reason": "Trainer evaluation failure after training is unrelated to Gemma4 device-map forward mismatches." + "reason": "Both mention tied weights, but one is a device-dependent serialization regression while the other is a broader translation-model tied-weights complaint; not clearly the same bug." }, { "left": "issue:30064", "right": "issue:33666", "accept": false, - "reason": "Void segmentation-map processing and Qwen2-VL multi-GPU training are unrelated bugs." + "reason": "Void segmentation map processing and Qwen2-VL multi-GPU training are unrelated image/model-training issues." }, { "left": "issue:44464", "right": "issue:44743", "accept": false, - "reason": "Both involve generation/cache behavior, but they are different mechanisms and different model families." + "reason": "Chunked generation with compiled forward and Qwen3 recurrent-state reset are different generation/runtime bugs." }, { "left": "issue:44361", "right": "issue:44442", "accept": false, - "reason": "Both are tokenizer failures, but they are different tokenizers with different failure modes." + "reason": "MLukeTokenizer AttributeError and FastSpeech2ConformerTokenizer loading failure are different tokenizer implementations and failures." }, { "left": "issue:43653", "right": "issue:44361", "accept": false, - "reason": "BigBird special-token decode behavior is unrelated to MLukeTokenizer task-time AttributeError failures." + "reason": "BigBird special-token decode issue vs MLukeTokenizer task AttributeError are unrelated tokenizer bugs." }, { "left": "issue:41720", "right": "issue:43906", "accept": false, - "reason": "The second is only an isolated reproduction of another issue; there is no evidence it is the same A800 cudaErrorAssert bug." + "reason": "Both reference model execution failures, but one is Qwen3 auto device mapping/cuda assert and the other is just a reproduction of an external issue." }, { "left": "issue:38175", "right": "issue:43295", "accept": false, - "reason": "Unexpected zero probabilities in SIGLIP2 is not the same as processor/tokenizer regression in v4.57.5." + "reason": "Zero probabilities in SigLIP2 vs processor.tokenizer/images regression are different model and API bugs." }, { "left": "issue:28282", "right": "issue:32090", "accept": false, - "reason": "A missing-PyTorch import error is unrelated to a Trainer distributed broadcast TypeError." + "reason": "ImportError for missing PyTorch and Trainer _gpu_broadcast_one NoneType error are unrelated." }, { "left": "issue:28282", "right": "issue:33357", "accept": false, - "reason": "Environment import failure and MacOS bus error on CLIP are different failure classes." + "reason": "Missing PyTorch import vs MacOS bus error in CLIP are different failure modes and environments." }, { "left": "issue:43653", "right": "issue:44568", "accept": false, - "reason": "Both involve special tokens, but they affect different tokenizers and different regressions." + "reason": "Both are tokenizer/special-token related, but BigBird mask-token registration and mdeberta BOS/EOS addition are distinct behaviors." }, { "left": "issue:43688", "right": "issue:45440", "accept": false, - "reason": "Aux-loss normalization in MoE models is not the same as DeepSeekV3 implementation divergence." + "reason": "Aux-loss normalization in MoE models and DeepSeekV3 implementation divergence are not the same code-path problem." }, { - "left": "issue:45491", - "right": "issue:46002", + "left": "issue:43577", + "right": "issue:43697", "accept": false, - "reason": "Both mention NaNs, but one is a sliding-window attention GPU issue and the other is a classification-head init/checkpoint issue." + "reason": "BLIP-2 dtype propagation and RTDetrV2 output drift under v5 are unrelated model-specific regressions." }, { - "left": "issue:43577", - "right": "issue:43697", + "left": "issue:45491", + "right": "issue:46002", "accept": false, - "reason": "A dtype-loading bug in BLIP2 is unrelated to RTDetrV2 output divergence in v5." + "reason": "Gemma3 NaN embeddings with mixed-length batching and classification-head NaNs after checkout/init are different initialization/attention issues." }, { "left": "issue:43688", "right": "issue:45704", "accept": false, - "reason": "MoE auxiliary-loss normalization and a T5 apex RMSNorm memory leak are unrelated." + "reason": "MoE auxiliary loss normalization and T5/apex FusedRMSNorm memory leak are unrelated bugs." }, { "left": "issue:28282", "right": "issue:35141", "accept": false, - "reason": "A PyTorch import error is not the same as output embeddings being reinitialized during post_init." + "reason": "PyTorch import failure and token-embedding resize/post_init reinitialization are different layers of the stack." }, { "left": "issue:44843", "right": "issue:44991", "accept": false, - "reason": "Offline model_info regression and a tokenizer-loading failure for a specific model are different bugs." + "reason": "Offline model_info call in AutoTokenizer and tokenizer loading failure for a specific model are not the same underlying defect." }, { "left": "issue:43688", "right": "issue:45701", "accept": false, - "reason": "Aux-loss normalization is unrelated to a general tokenization-change report." + "reason": "MoE auxiliary-loss normalization and a broad tokenization-change report are too different to be duplicates." }, { "left": "issue:43643", "right": "issue:44561", "accept": false, - "reason": "Both involve trust_remote_code breakage, but one is missing fields in AutoConfig output and the other is removal of a torch-fx utility." + "reason": "Missing fields from AutoConfig with trust_remote_code and removal of is_torch_fx_available breaking remote-code models are separate compatibility issues." }, { "left": "issue:43646", - "right": "issue:44561", - "accept": false, - "reason": "A broad custom-model initialization regression is not the same concrete bug as trust_remote_code breakage from removed APIs." - }, - { - "left": "issue:43653", - "right": "issue:44704", + "right": "issue:43749", "accept": false, - "reason": "A BigBird special-token decode bug is unrelated to AutoProcessor kwargs forwarding." + "reason": "Custom model initialization breakage in v5 and FSDP_CPU_RAM_EFFICIENT_LOADING failure are different migration/runtime problems." }, { "left": "issue:29159", "right": "issue:33357", "accept": false, - "reason": "Tokenizer inconsistency and a MacOS bus error are unrelated." + "reason": "Slow-vs-fast tokenizer inconsistency and MacOS CLIP bus error are unrelated." } ] }, @@ -6568,8 +6618,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -6657,232 +6707,220 @@ "issue:43720|issue:44361", "issue:28218|issue:33357" ], - "prepared_review_unit_hash": "dac03b1dc195efc491827ce99ff7e5b006bfd4fcb935003368e045e6a82ecb6a", + "prepared_review_unit_hash": "81ee92b2c9418d7f42d5d8f83cbc0758dda08a5e2aecac9e60b0cf617d593774", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 33, - "serialized_chars": 37648, - "estimated_input_tokens": 9412, - "estimated_eval_tokens": 19080 + "serialized_chars": 37639, + "estimated_input_tokens": 9410, + "estimated_eval_tokens": 19076 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a heterogeneous mix of unrelated Transformers issues spanning tokenization, config reloads, model loading, training, device/dtype mismatches, and generation regressions. None of the soft pairs look like true duplicates; the closest overall representative is the broad Granite tokenizer regression.", - "confidence": 0.89, - "canonical_issue_reason": "No single canonical issue: the items span many unrelated bugs and model families, so this is not a clean duplicate cluster.", + "summary": "This cluster is heterogeneous, but the strongest common thread is v5-era tokenizer/config regressions; most proposed soft matches are actually different model-specific loading/training/runtime bugs.", + "confidence": 0.84, + "canonical_issue_reason": "issue:43122 is the broadest tokenizer regression here: it describes a version-to-version tokenization change with the same tokenizer, which makes it the best umbrella issue for narrower tokenizer symptoms.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45812 is the strongest representative: it describes a broad silent v4\u2192v5 regression affecting all Granite models and has the highest inbound-references signal.", + "best_issue_reason": "issue:45812 is the strongest standalone issue in the set: it names a concrete, high-impact AutoTokenizer regression across a whole model family and already has inbound references.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43531", "right": "issue:43742", "accept": false, - "reason": "Different bugs: Qwen3-MoE sliding_window behavior vs MobileLLM loading key errors." + "reason": "Different bugs: Qwen3-MoE sliding-window behavior vs MobileLLM-125M load key error." }, { "left": "issue:43653", "right": "issue:43927", "accept": false, - "reason": "Tokenizer special-token registration bug vs config token-ID loss on save/load." + "reason": "Both involve token-related problems, but one is BigBird special-token registration and the other is DiaConfig save/load custom token IDs; not the same bug." }, { "left": "issue:44464", "right": "issue:44977", "accept": false, - "reason": "Both are generation bugs, but one is compiled forward chunking and the other is flash-attention/Qwen3.5-specific." + "reason": "Compiled-forward chunked generation inconsistency is separate from a flash-attention generation failure in Qwen3.5." }, { "left": "issue:43577", "right": "issue:45588", "accept": false, - "reason": "BLIP2 dtype persistence bug vs flash_attention crash on sink-less models; different code paths." + "reason": "Unrelated issues: dtype propagation in BLIP2 loading vs flash_attention.py crashing on s_aux=None." }, { "left": "issue:43550", "right": "issue:43927", "accept": false, - "reason": "torch.compile/SDPA failure for Bamba-9B-v2 is unrelated to DiaConfig token-ID persistence." + "reason": "Bamba torch.compile/SDPA failure is unrelated to DiaConfig token ID loss on save/load." }, { "left": "issue:44479", "right": "issue:44625", "accept": false, - "reason": "Video-input regression across Qwen VL variants vs num_labels propagation in Qwen3.5 config." + "reason": "Video-input regression for Qwen VL models is not the same as num_labels propagation between configs." }, { "left": "issue:43550", "right": "issue:44361", "accept": false, - "reason": "Bamba torch.compile/SDPA issue is unrelated to MLukeTokenizer task AttributeError." + "reason": "Different failure modes and models: Bamba compile/SDPA vs MLukeTokenizer AttributeError on tasks." }, { "left": "issue:43577", "right": "issue:44554", "accept": false, - "reason": "dtype loading mismatch in BLIP2 vs an MPS attention correctness issue." + "reason": "BLIP2 dtype handling and MPS attention correctness are unrelated code paths." }, { "left": "issue:43122", "right": "issue:43906", "accept": false, - "reason": "43906 is only an isolated reproduction reference; no evidence it is the same tokenizer regression as 43122." + "reason": "43906 is only an isolated reproduction reference; insufficient evidence that it is the same underlying tokenizer regression." }, { "left": "issue:41720", "right": "issue:44464", "accept": false, - "reason": "Different failures: auto device mapping cuda assert vs compiled-forward chunked generation inconsistency." + "reason": "Qwen3 auto device-map cuda assert and compiled-forward generation inconsistency are different bugs." }, { "left": "issue:32090", "right": "issue:34567", "accept": false, - "reason": "Trainer broadcast TypeError vs TrainerState token-count property not updating." + "reason": "Trainer GPU broadcast NoneType error and TrainerState token counter not updating are unrelated." }, { "left": "issue:43504", "right": "issue:43720", "accept": false, - "reason": "Legacy field loading failure in BeitForSemanticSegmentation vs BitNet packed-weight unpacking during accelerate loading." + "reason": "Legacy field loading in BeitForSemanticSegmentation is not the same as BitNet packed-weight unpacking during accelerate loading." }, { "left": "issue:43504", "right": "issue:43550", "accept": false, - "reason": "Legacy preset loading bug is unrelated to Bamba torch.compile/SDPA failure." + "reason": "Different model-loading paths and failures; no shared underlying defect is evident." }, { "left": "issue:43504", "right": "issue:44610", "accept": false, - "reason": "Different model-loader issue vs OmDet-Turbo processor image-size mismatch." + "reason": "Pretrained loading legacy-field issue is unrelated to OmDet-Turbo processor image-size mismatch." }, { "left": "issue:44466", "right": "issue:46050", "accept": false, - "reason": "Serialization of tied lm_head weights vs sinq quantized save/load breakage are not the same bug." + "reason": "Both touch serialization, but one is lm_head.weight device-dependent save behavior and the other is sinq quantization save/load breakage." }, { "left": "issue:43742", "right": "issue:44821", "accept": false, - "reason": "MobileLLM model loading failure vs AutoImageProcessor URL loading failure." + "reason": "MobileLLM key error and AutoImageProcessor-from-URL loading are different loader failures." }, { "left": "issue:32090", "right": "issue:33666", "accept": false, - "reason": "Trainer broadcast TypeError vs Qwen2-VL multi-GPU training are different training issues." + "reason": "Trainer broadcast error is unrelated to Qwen2-VL multi-GPU training support." }, { "left": "issue:33666", "right": "issue:36010", "accept": false, - "reason": "Multi-GPU training bug vs ImportError for GenerationMixin; unrelated failure modes." + "reason": "Multi-GPU training for Qwen2-VL and GenerationMixin import failure are unrelated." }, { "left": "issue:43452", "right": "issue:44451", "accept": false, - "reason": "gguf_file from_pretrained breakage is not the same as ScandiBERT load failure." + "reason": "gguf_file loading breakage is not the same as a specific model that cannot be loaded." }, { "left": "issue:45823", "right": "issue:46050", "accept": false, - "reason": "Gemma4 device-map mismatch vs quantized model save/load corruption are different paths." + "reason": "Gemma4 device-mapping mismatch is unrelated to sinq quantized save/load corruption." }, { "left": "issue:43278", "right": "issue:43688", "accept": false, - "reason": "Embedding dtype drift on eval vs auxiliary-loss normalization bug are unrelated." + "reason": "Embedding dtype drift during evaluate is unrelated to auxiliary-loss normalization in MoE models." }, { "left": "issue:44849", "right": "issue:45357", "accept": false, - "reason": "output_hidden_states regression in Qwen3.5 vs incorrect visual-encoder key saving are different bugs." + "reason": "Qwen3.5 hidden-states behavior and incorrect visual-encoder save keys are different regressions." }, { "left": "issue:43525", "right": "issue:43572", "accept": false, - "reason": "Both are missing-config-attribute reports, but they affect different models and different fields." + "reason": "Both mention missing pad-token fields, but they are different configs and different attribute names; not enough to treat as one bug." }, { "left": "issue:43606", "right": "issue:43927", "accept": false, - "reason": "CPU-offload device mismatch for bark-small vs DiaConfig token-ID loss are unrelated." + "reason": "CPU offload device mismatch and DiaConfig token-ID loss are unrelated." }, { "left": "issue:43720", "right": "issue:43927", "accept": false, - "reason": "BitNet packed-weight loading issue vs config token-ID persistence are different problems." + "reason": "Packed-weight unpacking during accelerate loading is unrelated to config token-ID persistence." }, { "left": "issue:43262", "right": "issue:43295", "accept": false, - "reason": "Audio processor chat-template sample-rate default vs processor.tokenizer/images regression are unrelated processor bugs." + "reason": "Audio chat-template sampling-rate default and processor/tokenizer API regression are separate processor issues." }, { "left": "issue:44625", "right": "issue:46002", "accept": false, - "reason": "num_labels propagation bug vs NaNs in classification heads upon init are different issues." + "reason": "num_labels propagation and NaNs in classification heads are different problems." }, { "left": "issue:43493", "right": "issue:43825", "accept": false, - "reason": "SigLIP2 implementation discrepancy vs pipeline translation-task error message are unrelated." + "reason": "SigLIP2 implementation discrepancy is unrelated to pipeline translation-task messaging." }, { "left": "issue:44568", "right": "issue:45812", "accept": false, - "reason": "mDeBERTa special-token addition regression vs Granite token-ID regression are both tokenizer bugs but not the same underlying change." - }, - { - "left": "issue:43756", - "right": "issue:44913", - "accept": false, - "reason": "Smollm3 RoPE-layer mismatch vs GPTNeoXConfig rotary_pct reload bug are different config/model issues." - }, - { - "left": "issue:33666", - "right": "issue:36331", - "accept": false, - "reason": "Qwen2-VL multi-GPU training vs CustomTrainer compute_loss signature mismatch." + "reason": "Both are tokenization regressions, but one is mdeberta special-token insertion and the other is Granite token ID mapping; different concrete bugs." }, { "left": "issue:43720", "right": "issue:44361", "accept": false, - "reason": "BitNet packed-weight loading vs MLukeTokenizer task AttributeError are unrelated." + "reason": "Different loading paths and failures: BitNet accelerate unpacking vs MLukeTokenizer AttributeError." }, { "left": "issue:28218", "right": "issue:33357", "accept": false, - "reason": "Tokenizer added-space behavior vs MacOS bus error on CLIP model are unrelated bugs." + "reason": "Tokenizer added-space behavior and MacOS CLIP bus error are unrelated." } ] }, @@ -6901,8 +6939,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -6965,14 +7003,14 @@ "issue:45803|issue:45987", "issue:45200|issue:45229", "issue:44488|issue:44843", - "issue:43257|issue:43299", "issue:43756|issue:45083", + "issue:43257|issue:43299", "issue:43334|issue:45759", "issue:39692|issue:43295", "issue:43824|issue:44589", "issue:45229|issue:45823", - "issue:43646|issue:44466", "issue:43572|issue:44291", + "issue:43646|issue:44466", "issue:44589|issue:45520", "issue:33357|issue:33666", "issue:43504|issue:43606", @@ -6987,14 +7025,14 @@ "issue:44423|issue:44734", "issue:45406|issue:46121" ], - "prepared_review_unit_hash": "6768b82f893b5ebff69e53c1c4b173d8b6a6c9972c7b40fe56ef00a62b2615ad", + "prepared_review_unit_hash": "4c7e0bfe170f5a055da0aaa5fd766cf4bde154a0ddf1f1b42f7753c796411ba5", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, @@ -7009,198 +7047,192 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All suggested soft edges are false positives: the paired issues share broad vocabulary or a subsystem, but they describe different concrete bugs, models, or code paths.", - "confidence": 0.96, - "canonical_issue_reason": "No single canonical issue stands out; the items are heterogeneous and the soft pairs are only topical overlaps, not duplicates.", - "canonical_pr_reason": null, - "best_issue_reason": "No single best issue for duplicate-canonicalization: the set spans unrelated failures across different models, configs, and runtime paths.", - "best_pr_reason": null, + "summary": "This cluster is heterogeneous: most items are unrelated model-loading, config, serving, or training bugs, and the soft pairs are topical similarities rather than true duplicates. No single issue or PR cleanly represents the cluster.", + "confidence": 0.99, + "canonical_issue_reason": "No canonical issue: the items span many distinct models and failure modes, so there is no shared underlying bug to collapse onto one issue.", + "canonical_pr_reason": "No PRs are present in the cluster.", + "best_issue_reason": "No global best issue: any one item would be too narrow and would misrepresent the cluster\u2019s mixed bug themes.", + "best_pr_reason": "No PRs are present in the cluster.", "soft_edge_verdicts": [ { "left": "issue:44964", "right": "issue:45405", "accept": false, - "reason": "One is a model-loading failure for Phi-4 multimodal; the other is a dependency-version bump for PEFT. Unrelated root causes." + "reason": "Both are loading-related, but one is a Phi-4 multimodal model load failure and the other is a PEFT version pin/release mismatch; different root causes." }, { "left": "issue:39692", "right": "issue:43493", "accept": false, - "reason": "Docs example mistakes for SigLIP2 are not the same as a HF-vs-JAX implementation discrepancy." + "reason": "Same model family, but one is a documentation/example problem while the other is an HF-vs-JAX implementation discrepancy; not the same bug." }, { "left": "issue:43540", "right": "issue:44977", "accept": false, - "reason": "Qwen3OmniMoe video-input validation and Qwen3.5 flash-attention generation issues are different code paths and failure modes." + "reason": "Both involve Qwen multimodal models, but the failures are different: video input validation in Qwen3OmniMoe vs flash-attention generation behavior in Qwen3.5." }, { "left": "issue:43493", "right": "issue:43844", "accept": false, - "reason": "SigLIP2 implementation mismatch is unrelated to ZeRO-3 gradient growth during training." + "reason": "SigLIP2 discrepancy vs DeepSpeed gradient explosion are different code paths and symptoms; no shared underlying defect." }, { "left": "issue:43278", "right": "issue:43844", "accept": false, - "reason": "Embedding dtype drift at eval time is not the same bug as abnormal gradient increase under DeepSpeed." + "reason": "Embedding dtype regression during eval is unrelated to the ZeRO-3 gradient anomaly." }, { "left": "issue:43334", "right": "issue:43572", "accept": false, - "reason": "Both mention missing config fields, but they affect different models and different attributes (`pad_token_id` vs `pad_token_idx`)." + "reason": "Both are config attribute issues, but they affect different models and missing fields; not the same problem." }, { "left": "issue:45803", "right": "issue:45987", "accept": false, - "reason": "Bare `except:` in Fuyu tensor conversion and a cache-length return-type contract bug are unrelated." + "reason": "Both touch utility contracts, but one is a bare-except bug in Fuyu tensor conversion and the other is a StaticCache return-type mismatch." }, { "left": "issue:45200", "right": "issue:45229", "accept": false, - "reason": "Defaulting `mm_token_type_ids` to zeros and a Gemma4 multi-GPU OOM are distinct problems." + "reason": "Gemma4 mm_token_type_ids defaulting and Gemma4 multi-GPU OOM are separate issues with different failure modes." }, { "left": "issue:44488", "right": "issue:44843", "accept": false, - "reason": "A specific pretrained model load failure is not the same as offline-mode breakage from an unconditional hub call." + "reason": "A model load failure and an offline-mode tokenizer hub call are only loosely related; different underlying causes." }, { - "left": "issue:43257", - "right": "issue:43299", + "left": "issue:43756", + "right": "issue:45083", "accept": false, - "reason": "Both involve Qwen3 MoE, but one is accelerate/deepspeed weight conversion and the other is a multimodal loading regression." + "reason": "Both mention RoPE-related internals, but one is about missing RoPE layers in Smollm3 and the other is a helper length computation in qwen3_omni_moe." }, { - "left": "issue:43756", - "right": "issue:45083", + "left": "issue:43257", + "right": "issue:43299", "accept": false, - "reason": "Smollm3 RoPE-layer retention and a qwen3_omni_moe helper output-length issue are unrelated." + "reason": "Both concern Qwen3 MOE loading, but one is accelerate+deepspeed weight conversion and the other is a dev-version load regression; not the same concrete bug." }, { "left": "issue:43334", "right": "issue:45759", "accept": false, - "reason": "A missing `pad_token_id` on Qwen3-VL is separate from `from_config` failing to unwrap `text_config` in composite Qwen configs." + "reason": "Both are multimodal config issues, but one is missing pad_token_id on Qwen3-VL text config and the other is failure to unwrap text_config in AutoModelForCausalLM.from_config." }, { "left": "issue:39692", "right": "issue:43295", "accept": false, - "reason": "SigLIP2 example/documentation errors do not match the processor/tokenizer regression in custom model code." + "reason": "Documentation/example errors for SigLIP2 are unrelated to the processor.tokenizer regression in a separate multimodal code path." }, { "left": "issue:43824", "right": "issue:44589", "accept": false, - "reason": "A missing Qwen2.5-VL import and a Float8 storage lookup error are unrelated." + "reason": "ImportError for Qwen2_5_VL and missing Float8 storage support are unrelated issues." }, { "left": "issue:45229", "right": "issue:45823", "accept": false, - "reason": "Both mention Gemma4 and devices, but one is CUDA OOM under multi-GPU inference and the other is a device mismatch in forward." - }, - { - "left": "issue:43646", - "right": "issue:44466", - "accept": false, - "reason": "Custom model initialization breakage and inconsistent `lm_head.weight` serialization are different lifecycle bugs." + "reason": "Both are Gemma4-related, but one is an inference OOM and the other is a device mismatch with device_map='auto'; different bugs." }, { "left": "issue:43572", - "right": "issue:45405", - "accept": false, - "reason": "StableLmConfig missing `pad_token_idx` has nothing to do with an unreleased PEFT version bump." - }, - { - "left": "issue:43994", - "right": "issue:44493", - "accept": false, - "reason": "SigLIP2 bad outputs with AutoModel/pipeline and unexpected position-id keys in many models are not the same issue." - }, - { - "left": "issue:44792", - "right": "issue:45405", + "right": "issue:44291", "accept": false, - "reason": "A Janus image-generation test failure is unrelated to the PEFT version pin problem." + "reason": "StableLmConfig missing pad_token_idx and init_empty_weights receiving _is_hf_initialized are distinct initialization/config regressions." }, { - "left": "issue:43278", - "right": "issue:43701", + "left": "issue:43646", + "right": "issue:44466", "accept": false, - "reason": "Eval-time dtype drift and resume-from-checkpoint key mismatch are different bugs." + "reason": "Custom model initialization breakage and lm_head serialization inconsistency are different bugs in different parts of the stack." }, { - "left": "issue:44423", - "right": "issue:44734", + "left": "issue:44589", + "right": "issue:45520", "accept": false, - "reason": "Both are in `serve`, but continuous-batching multimodal crashes and KV-cache continuation indexing failures are separate code paths." + "reason": "Float8 storage lookup failure and flash_attn import_utils KeyError are unrelated environment/import problems." }, { - "left": "issue:45406", - "right": "issue:46121", + "left": "issue:33357", + "right": "issue:33666", "accept": false, - "reason": "Gemma4Processor missing `_tokenizer` in serve is unrelated to rope-validation list handling." + "reason": "MacOS bus error with CLIP and Qwen2-VL multi-GPU training are completely different issues." }, { "left": "issue:43504", "right": "issue:43606", "accept": false, - "reason": "A legacy-field load failure for Beit semantic segmentation is unrelated to a CPU-offload device mismatch in bark-small." + "reason": "BEiT legacy field loading and bark-small CPU offload device mismatch are unrelated model-specific bugs." }, { "left": "issue:43927", "right": "issue:44361", "accept": false, - "reason": "Custom token IDs being lost on save/load is not the same as MLukeTokenizer throwing an AttributeError on tasks." + "reason": "DiaConfig token ID persistence and MLukeTokenizer AttributeError affect different components and error modes." }, { "left": "issue:44464", "right": "issue:45584", "accept": false, - "reason": "Chunked generation with compiled forward and Whisper empty-transcription handling are different features and bugs." + "reason": "Chunked generation inconsistency with compiled forward is not the same as Whisper empty-transcription failure after align_special_tokens." }, { "left": "issue:43504", "right": "issue:43653", "accept": false, - "reason": "Beit legacy-field loading and BigBirdTokenizer special-token registration are unrelated." + "reason": "Both are loading/tokenizer-related, but BEiT legacy-field loading and BigBirdTokenizer special-token registration are separate bugs." }, { "left": "issue:43606", "right": "issue:44112", "accept": false, - "reason": "A real device-mismatch bug in bark-small is not the same as a stale GraniteSpeech CI test." + "reason": "A real CPU-offload device mismatch is not the same as a stale CI test failure." }, { "left": "issue:43575", "right": "issue:45405", "accept": false, - "reason": "TP loading OOM for Qwen2-57B is unrelated to the PEFT release-version issue." + "reason": "Qwen2-57B TP OOM and a PEFT version bump are unrelated." }, { "left": "issue:43994", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 nonsensical outputs and unexpected position-id keys are different symptoms with different likely fixes." + "reason": "SigLIP2 producing nonsensical outputs and unexpected position-id keys across many models are different regression classes." }, { "left": "issue:44792", "right": "issue:45405", "accept": false, - "reason": "Janus image-generation test failure and PEFT version pinning are unrelated." + "reason": "A janus test failure and a PEFT minimum-version bump are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype regression and resume_from_checkpoint key mismatch do not share a concrete code path." }, { "left": "issue:44423", - "right": "issue:45584", + "right": "issue:44734", + "accept": false, + "reason": "Both are serving-time crashes, but one is a multimodal 'str.to' error and the other is KV-cache tensor indexing; distinct bugs." + }, + { + "left": "issue:45406", + "right": "issue:46121", "accept": false, - "reason": "Serving multimodal continuous batching and Whisper empty-output after token alignment are separate bugs." + "reason": "Gemma4Processor missing _tokenizer and JSON-loaded rope-key validation TypeError are unrelated." } ] }, @@ -7219,8 +7251,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -7298,221 +7330,221 @@ "issue:43572|issue:44987", "issue:43572|issue:43957", "issue:37428|issue:43824", + "issue:43295|issue:43450", "issue:44898|issue:45412", "issue:45200|issue:45325", - "issue:43295|issue:43450", "issue:44466|issue:45127", "issue:43844|issue:44936", "issue:43295|issue:43873" ], - "prepared_review_unit_hash": "9ff21b011390503f3a829a16f4f03b5f69b1e9c141de5596c129145e86ff2ec8", + "prepared_review_unit_hash": "3f97a6ca6c30821392fbd89d94cc395c5194aac6667faf205f373f6deb694b3b", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36784, - "estimated_input_tokens": 9196, - "estimated_eval_tokens": 18648 + "serialized_chars": 36785, + "estimated_input_tokens": 9197, + "estimated_eval_tokens": 18650 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This set is not a single duplicate cluster; it contains many unrelated issues, and all listed soft pairs look like false-positive similarity matches rather than the same underlying bug.", - "confidence": 0.94, - "canonical_issue_reason": "No canonical issue emerges; the items span unrelated tokenizer, generation, vision, quantization, trainer, and import-path bugs.", + "summary": "The set is a mixed bag of unrelated Hugging Face/Transformers bug reports; none of the soft pairs are strong enough to merge as duplicates. There are no PRs in the cluster. If a representative issue is needed, issue 44521 is the strongest single candidate because it is specific, reproducible, and has the most substantive discussion among the active reports.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 44521 is the clearest, most actionable open bug report here: it describes a concrete multimodal `apply_chat_template` failure, has strong discussion/inbound references, and is more specific than the broader or more ambiguous issues.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good global representative for deduplication here, because none of the soft-edge pairs appears to describe the same concrete failure.", + "best_issue_reason": "Issue 44521 is the best representative issue overall because it is well-scoped, user-facing, and has enough engagement to serve as a stable canonical reference.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43873", "right": "issue:46050", "accept": false, - "reason": "Quantization/offloading behavior vs quantized save/load breakage; related theme, different failure mode and code path." + "reason": "Both mention quantization, but one is an offloading behavior issue and the other is a save/load corruption regression; different code paths and symptoms." }, { "left": "issue:29942", "right": "issue:30333", "accept": false, - "reason": "Flash Attention 2 test failures and MLFlow reporting on failed jobs are unrelated subsystems." + "reason": "FlashAttention CI failures and MLflow reporting behavior are unrelated subsystems." }, { "left": "issue:45588", "right": "issue:46050", "accept": false, - "reason": "A flash_attention.py None handling crash is not the same bug as quantized model save/load corruption." + "reason": "A flash-attention crash on `s_aux=None` is unrelated to a quantized model breaking after save/load." }, { "left": "issue:42490", "right": "issue:43696", "accept": false, - "reason": "Local-path import warning and CUDA OOM are completely different issues." + "reason": "A local-path import warning and a CUDA OOM report are not the same bug." }, { "left": "issue:44521", "right": "issue:45290", "accept": false, - "reason": "Both involve chat templates, but one is zero assistant masks for multimodal inputs and the other crashes on tool-call assistant messages." + "reason": "Both involve `apply_chat_template`, but one is all-zero assistant masks for multimodal inputs while the other crashes on tool-call assistant messages with no content." }, { "left": "issue:43844", "right": "issue:44964", "accept": false, - "reason": "ZeRO-3 gradient growth and Phi-4 multimodal loading failure are unrelated." + "reason": "ZeRO-3 gradient instability and a Phi-4 multimodal loading failure are unrelated." }, { "left": "issue:28218", "right": "issue:34567", "accept": false, - "reason": "Added-token spacing and TrainerState token counting are separate tokenizer/training issues." + "reason": "Tokenizer spacing after added tokens is unrelated to TrainerState token-count tracking." }, { "left": "issue:45084", "right": "issue:45520", "accept": false, - "reason": "Template compilation error and flash_attn import KeyError are unrelated." + "reason": "A compilation error for non-template nodes is not the same as a Python 3.13 `flash_attn` import key error." }, { "left": "issue:44554", "right": "issue:45446", "accept": false, - "reason": "MPS attention correctness and a PyTorch version check for AuxRequest are different code paths." + "reason": "An MPS attention correctness issue and a PyTorch version-check/import guard bug are different problems." }, { "left": "issue:29159", "right": "issue:32090", "accept": false, - "reason": "Slow/fast tokenizer inconsistency is unrelated to the NoneType broadcast error in Trainer." + "reason": "Slow/fast tokenizer inconsistency is unrelated to the `_gpu_broadcast_one` NoneType error in Trainer." }, { "left": "issue:43540", "right": "issue:45200", "accept": false, - "reason": "Video input processing in Qwen3OmniMoe is not the same as Gemma4 mm_token_type_ids defaults." + "reason": "Video-input processing in Qwen3OmniMoe is unrelated to Gemma 4 text-only `mm_token_type_ids` defaults." }, { "left": "issue:43582", "right": "issue:44977", "accept": false, - "reason": "Apple Silicon allocator TypeError and Qwen3.5 flash-attention generation are unrelated." + "reason": "Apple Silicon allocator warmup errors and Qwen3.5 flash-attention generation failures are different code paths." }, { "left": "issue:15354", "right": "issue:38175", "accept": false, - "reason": "TorchScript GeneratorExp export failure and SigLIP2 zero probabilities are distinct problems." + "reason": "TorchScript GeneratorExp export failure is unrelated to the SigLIP2 zero-probability report." }, { "left": "issue:43232", "right": "issue:44849", "accept": false, - "reason": "Generation kwarg update after sync_gpus is unrelated to output_hidden_states behavior in Qwen3.5." + "reason": "A generation kwargs sync issue is not the same as the Qwen3.5 `output_hidden_states=True` bug." }, { "left": "issue:37428", "right": "issue:43957", "accept": false, - "reason": "Flash-attention import error and meta-device model loading breakage are different regressions." + "reason": "A flash-attention import symbol error and a `torch.device('meta')` model-loading regression are distinct failures." }, { "left": "issue:45200", "right": "issue:45823", "accept": false, - "reason": "Defaulting mm_token_type_ids to zeros and a device_map forward mismatch in Gemma4 are not the same bug." + "reason": "Missing default multimodal token type ids and a Gemma4 PLE device-mapping mismatch are not the same bug." }, { "left": "issue:45072", "right": "issue:45198", "accept": false, - "reason": "bfloat16 dtype mismatches in CI and Wav2Vec2 save/tokenization failure are unrelated." + "reason": "Dtype mismatches in inference and Wav2Vec2 save/tokenization failures do not share a concrete code-path." }, { "left": "issue:44315", "right": "issue:44849", "accept": false, - "reason": "Liger Kernel application in model_init and Qwen3.5 hidden-states output are unrelated." + "reason": "Liger Kernel not being applied with `model_init` is unrelated to the hidden-states bug in Qwen3.5." }, { "left": "issue:43493", "right": "issue:43701", "accept": false, - "reason": "SigLIP2 implementation discrepancy and resume_from_checkpoint key mismatch do not share a concrete failure path." + "reason": "A SigLIP2 HF/JAX discrepancy and a resume-checkpoint key mismatch are different issues." }, { "left": "issue:43295", "right": "issue:44610", "accept": false, - "reason": "Processor.tokenizer regression and OmDet-Turbo image-size mismatch are different processor bugs." + "reason": "The processor/tokenizer regression and the OmDet-Turbo image-size mismatch are separate preprocessing bugs." }, { "left": "issue:44554", "right": "issue:45356", "accept": false, - "reason": "MPS attention correctness and Kimi tokenizer codec regression are unrelated." + "reason": "An MPS attention correctness problem is unrelated to the Kimi-K2.5 tokenizer regression." }, { "left": "issue:43572", "right": "issue:44987", "accept": false, - "reason": "StableLmConfig pad_token_idx and loading physical-intelligence/fast are distinct loading/configuration issues." + "reason": "A missing `pad_token_idx` config field is not the same as a model-loading failure for `physical-intelligence/fast`." }, { "left": "issue:43572", "right": "issue:43957", "accept": false, - "reason": "Both are v5 loading regressions, but the config default bug is not the same as meta-device loading failures." + "reason": "Config defaults and meta-device loading failures are different regression types." }, { "left": "issue:37428", "right": "issue:43824", "accept": false, - "reason": "Missing flash_attention_utils symbol and missing Qwen2_5_VL export are separate import errors." + "reason": "These are both import errors, but they involve different symbols and model families with no shared concrete bug." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "A processor/tokenizer regression is not the same as batched video shape handling." }, { "left": "issue:44898", "right": "issue:45412", "accept": false, - "reason": "Perceiver resize failure and RT-DETR memory release are unrelated." + "reason": "Perceiver interpolation failures and RT-DETR memory-release behavior are unrelated." }, { "left": "issue:45200", "right": "issue:45325", "accept": false, - "reason": "Gemma4 token_type_ids defaults and Qwen2.5-VL rope-index scaling are different multimodal bugs." - }, - { - "left": "issue:43295", - "right": "issue:43450", - "accept": false, - "reason": "Processor.tokenizer regression and batched video-shape bug are not the same underlying issue." + "reason": "Gemma 4 text-only defaults and Qwen2.5-VL rope-index scaling are different multimodal bugs." }, { "left": "issue:44466", "right": "issue:45127", "accept": false, - "reason": "Both mention tied weights, but one is device-dependent lm_head serialization and the other is LoRA merge collapse with extended vocab; not the same code-path bug." + "reason": "Both touch tied weights, but one is serialization/device-dependent state and the other is LoRA merge collapse with extended vocab; too different to merge." }, { "left": "issue:43844", "right": "issue:44936", "accept": false, - "reason": "ZeRO-3 gradient anomaly and trainer.evaluate failing after train are different training bugs." + "reason": "Gradient explosion during training and `trainer.evaluate()` failing after `trainer.train()` are not the same failure mode." }, { "left": "issue:43295", "right": "issue:43873", "accept": false, - "reason": "Custom processor.tokenizer regression and quantization/offloading behavior are unrelated." + "reason": "Processor/tokenizer image handling and quantization offloading are unrelated subsystems." } ] }, @@ -7531,8 +7563,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -7593,14 +7625,14 @@ "issue:42915|issue:43824", "issue:43975|issue:45520", "issue:42490|issue:43493", - "issue:43278|issue:43856", "issue:28218|issue:35141", + "issue:43278|issue:43856", "issue:38175|issue:43493", - "issue:38175|issue:45412", - "issue:43653|issue:44112", "issue:43450|issue:43575", - "issue:44514|issue:44521", + "issue:43653|issue:44112", + "issue:38175|issue:45412", "issue:44442|issue:45198", + "issue:44514|issue:44521", "issue:44206|issue:45584", "issue:43827|issue:43901", "issue:43262|issue:43450", @@ -7616,218 +7648,209 @@ "issue:44448|issue:44743", "issue:43278|issue:43825" ], - "prepared_review_unit_hash": "da7cbb22dddc3e94dbf3b9c31e4c782cf95c9c158eeb8c22375565aa0be9dcb8", + "prepared_review_unit_hash": "822e6bbf16e338a2c1e5a88b2f3efb4b099947d4b778664de1cf8c87542e85e6", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36533, - "estimated_input_tokens": 9134, - "estimated_eval_tokens": 18524 + "serialized_chars": 36537, + "estimated_input_tokens": 9135, + "estimated_eval_tokens": 18526 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Most items are unrelated standalone issues. The only soft edge I\u2019d treat as a likely duplicate is the SigLIP2 pair; the rest are too model- or code-path-specific to merge safely.", - "confidence": 0.62, - "canonical_issue_reason": "issue:43493 is the broadest and most representative bug report in the one likely duplicate pair, covering the SigLIP2 implementation discrepancy that can explain the zero-probability symptom.", + "summary": "The soft links are mostly superficial title similarity across unrelated bugs. I would not merge any of the proposed pairs; they span different models, different APIs, or different failure modes. The best issue representative is the open multimodal chat-template bug because it is user-facing, active, and has the strongest issue-level signal among this noisy set.", + "confidence": 0.84, + "canonical_issue_reason": "issue:44521 is the most representative issue here: it is open, has the strongest inbound-reference/discussion signal, and describes a concrete user-facing bug in a core multimodal API (`apply_chat_template`).", "canonical_pr_reason": null, - "best_issue_reason": "issue:43493 is the clearest standalone representative of the cluster\u2019s closest match and has the most general wording among the likely duplicates.", + "best_issue_reason": "issue:44521 is the best standalone issue candidate for the same reasons: it is active, specific, and broader/more actionable than the other candidates in this set.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43334", "right": "issue:43525", "accept": false, - "reason": "Same symptom class (missing pad_token_id), but different config/model families and likely separate fixes." + "reason": "Both mention missing `pad_token_id`, but they affect different model configs (`Qwen3VLTextConfig` vs `Llama4Config`) and do not clearly describe the same concrete code-path bug." }, { "left": "issue:44792", "right": "issue:45563", "accept": false, - "reason": "Different generate-related failures; one is a model test, the other is a stale warning." + "reason": "These are unrelated failures: a Janus image-generation test failure versus a stale `num_return_sequences` warning in paged generation." }, { "left": "issue:43824", "right": "issue:43975", "accept": false, - "reason": "ImportError for a missing class versus a detokenization regression; not the same bug." + "reason": "One is an import error for a missing top-level class; the other is a detokenization regression for a specific model. Different bugs, different code paths." }, { "left": "issue:43388", "right": "issue:43526", "accept": false, - "reason": "Both concern labels, but in different components and with different failure modes." + "reason": "Both involve labels, but one is `gather_for_metrics` dropping tuple elements while the other is `BeitImageProcessorFast.reduce_labels` returning the wrong label array. Not the same bug." }, { "left": "issue:42915", "right": "issue:43824", "accept": false, - "reason": "Different subsystems and symptoms: FP8 config failure versus VL class import error." + "reason": "A Qwen3MoE FP8 config failure and a missing-export import error are unrelated." }, { "left": "issue:43975", "right": "issue:45520", "accept": false, - "reason": "Unrelated issues: detokenization in v5 versus flash-attn import handling on Python 3.13." + "reason": "Detokenization for DeepSeek-Coder and a Python 3.13 `flash_attn` import-utils error are unrelated." }, { "left": "issue:42490", "right": "issue:43493", "accept": false, - "reason": "Local-path import warning is unrelated to SigLIP2 output correctness." + "reason": "A local-path import warning and a SigLIP2 correctness discrepancy are not the same underlying issue." }, { - "left": "issue:43278", - "right": "issue:43856", + "left": "issue:28218", + "right": "issue:35141", "accept": false, - "reason": "Evaluate dtype drift and MoE memory usage are different bugs." + "reason": "Tokenizer spacing after added tokens and embedding reinitialization after resizing are separate tokenizer/model-init bugs." }, { - "left": "issue:28218", - "right": "issue:35141", + "left": "issue:43278", + "right": "issue:43856", "accept": false, - "reason": "Tokenizer spacing bug and embedding reinit bug are distinct code paths." + "reason": "BF16/FP32 dtype drift in evaluation and Qwen3 MoE memory inefficiency are different training/runtime problems." }, { "left": "issue:38175", "right": "issue:43493", - "accept": true, - "reason": "Both are SigLIP2 output-correctness reports; 38175 looks like a concrete symptom of the broader implementation discrepancy in 43493." + "accept": false, + "reason": "Both mention SigLIP2, but one is zero probabilities and the other is an HF-vs-JAX discrepancy; not enough to conclude the same bug." }, { - "left": "issue:38175", - "right": "issue:45412", + "left": "issue:43450", + "right": "issue:43575", "accept": false, - "reason": "SigLIP2 output discrepancy versus RT-DETR memory cleanup; unrelated." + "reason": "Batched video-shape handling and tensor-parallel OOM on a large model are unrelated." }, { "left": "issue:43653", "right": "issue:44112", "accept": false, - "reason": "Tokenizer special-token registration and GraniteSpeech CI failure are unrelated." - }, - { - "left": "issue:43450", - "right": "issue:43575", - "accept": false, - "reason": "Video input shape bug and tensor-parallel OOM are different problems." + "reason": "A tokenizer special-token registration bug and a GraniteSpeech CI test issue are not the same defect." }, { - "left": "issue:44514", - "right": "issue:44521", + "left": "issue:38175", + "right": "issue:45412", "accept": false, - "reason": "Same processor family, but batched-padding crash and zero assistant masks are distinct failures." + "reason": "Both are model/runtime complaints, but one is SigLIP2 output behavior and the other is RT-DETR memory release; unrelated." }, { "left": "issue:44442", "right": "issue:45198", "accept": false, - "reason": "Different tokenizer/load paths and different model families." + "reason": "AutoTokenizer loading failure and Wav2Vec2 save/tokenization failure are different tokenizer/model integration bugs." }, { - "left": "issue:43819", - "right": "issue:44857", + "left": "issue:44514", + "right": "issue:44521", "accept": false, - "reason": "DAC latent reconstruction mismatch and LwDetr AMP crash are unrelated." + "reason": "Both involve `apply_chat_template`, but one crashes on batched input with `padding=False` while the other returns all-zero assistant masks for multimodal inputs; these are not clearly the same concrete bug." }, { "left": "issue:44206", "right": "issue:45584", "accept": false, - "reason": "Feature extractor argument mismatch versus Whisper empty-transcription handling." + "reason": "An unsupported `center` argument in an audio feature extractor and Whisper failing on empty transcription are unrelated." }, { "left": "issue:43827", "right": "issue:43901", "accept": false, - "reason": "Both are docs issues, but they refer to different pipeline behaviors and different deprecations." + "reason": "Both are docs mismatches, but they concern different APIs and different deprecations (`pipeline()` removal vs `return_all_scores`)." }, { "left": "issue:43262", "right": "issue:43450", "accept": false, - "reason": "Audio sampling-rate default bug and batched video-shape bug are unrelated." + "reason": "Audio chat-template sampling-rate defaults and video batched-shape handling are unrelated processor bugs." }, { "left": "issue:43819", "right": "issue:44112", "accept": false, - "reason": "Unrelated issues in different models and code paths." + "reason": "A DAC latent/STE mismatch and a stale GraniteSpeech test are unrelated." }, { "left": "issue:44610", "right": "issue:44857", "accept": false, - "reason": "Different model processor/loss failures with no shared underlying bug." + "reason": "Processor output resolution mismatch and an AMP crash in a loss implementation are different issues." }, { "left": "issue:43819", "right": "issue:45083", "accept": false, - "reason": "DAC forward/latent mismatch versus qwen3_omni feature-length helper behavior." + "reason": "DAC latent reconstruction mismatch and a Qwen3 Omni helper length calculation bug are unrelated." }, { "left": "issue:37428", "right": "issue:43425", "accept": false, - "reason": "Missing internal import and general torch version incompatibility are not the same issue." + "reason": "A missing flash-attention utility import and Torch 2.10 incompatibility are not the same bug." }, { "left": "issue:45127", "right": "issue:45245", "accept": false, - "reason": "LoRA merge collapse and category-count limit are unrelated runtime problems." + "reason": "LoRA merge collapse with extended vocab and a category-count runtime limit are unrelated." }, { "left": "issue:43335", "right": "issue:43550", "accept": false, - "reason": "SwitchTransformers config bug and torch.compile/SDPA failure are different issues." + "reason": "A sparse-layer config bug and a torch.compile/SDPA failure in Bamba are different model/runtime problems." }, { "left": "issue:43606", "right": "issue:45083", "accept": false, - "reason": "CPU-offload device mismatch and qwen3_omni helper behavior are unrelated." + "reason": "CPU-offload device mismatch and a Qwen3 Omni feature-length helper bug are unrelated." }, { "left": "issue:44857", "right": "issue:45593", "accept": false, - "reason": "Different training bugs affecting different models and code paths." + "reason": "One is an AMP/CUDA crash in loss computation; the other is missing auxiliary losses when denoising is off. Different code paths." }, { "left": "issue:45290", "right": "issue:45381", "accept": false, - "reason": "Chat-template tool-call crash and Qwen2.5-VL vision-position-id bug are separate failures." + "reason": "`apply_chat_template` tool-call handling and Qwen2.5-VL video position IDs are different multimodal bugs." }, { - "left": "issue:43278", - "right": "issue:43825", + "left": "issue:44448", + "right": "issue:44743", "accept": false, - "reason": "Embedding dtype regression and pipeline v5 translation messaging are unrelated." + "reason": "Pegasus v4/v5 output differences and Qwen3 modular recurrent-state reset issues are unrelated." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "Grounded overall. The summary stays conservative, and the only accepted soft edge (38175 \u2194 43493) is plausibly the same SigLIP2 bug class from the packet. The other pairwise rejections are appropriately cautious given only title-level 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 }, @@ -7839,8 +7862,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -7902,10 +7925,10 @@ "issue:43335|issue:43653", "issue:43550|issue:44112", "issue:43819|issue:44493", - "issue:44360|issue:44561", "issue:44855|issue:45083", - "issue:39692|issue:43526", + "issue:44360|issue:44561", "issue:43335|issue:43504", + "issue:39692|issue:43526", "issue:45310|issue:45478", "issue:45588|issue:45859", "issue:43335|issue:45083", @@ -7914,273 +7937,266 @@ "issue:44292|issue:45520", "issue:43720|issue:44112", "issue:44448|issue:44779", - "issue:43335|issue:45072", "issue:43335|issue:44112", + "issue:43335|issue:45072", "issue:38175|issue:44493", "issue:44977|issue:45229", "issue:42490|issue:43854", "issue:43606|issue:44857", "issue:34689|issue:35141", - "issue:40444|issue:43526", "issue:44857|issue:45072", + "issue:40444|issue:43526", "issue:44561|issue:44779", - "issue:44462|issue:44561", "issue:44448|issue:44977", + "issue:44462|issue:44561", "issue:43723|issue:44554", "issue:28218|issue:32090", - "issue:43701|issue:45083", - "issue:43550|issue:45083" + "issue:43701|issue:45083" ], - "prepared_review_unit_hash": "cdbf0afc9934e80c812915a6beacd1ce2321e1e8498c3b0f174966f28f93f2b2", + "prepared_review_unit_hash": "e3e1be07932ce3d606471ff27e17f5c2767514a7241eafaf293c9b5a91cea8cb", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 36, - "serialized_chars": 38458, - "estimated_input_tokens": 9615, - "estimated_eval_tokens": 19486 + "soft_pair_count": 35, + "serialized_chars": 38156, + "estimated_input_tokens": 9539, + "estimated_eval_tokens": 19334 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly heterogeneous. The only clear duplicate pair is the Qwen3.5 MoE `from_pretrained` regression in issues 45310 and 45478; the rest describe unrelated bugs across different models/code paths.", - "confidence": 0.95, - "canonical_issue_reason": "Issue 45310 is the best canonical issue because it is the earlier report of the same Qwen3.5 MoE `from_pretrained` failure that 45478 repeats.", + "summary": "This cluster is mostly a mixed bag of unrelated issues. The only clear duplicate pair is the Qwen3.5 Moe `from_pretrained` error reports; the rest share broad themes but not the same underlying bug.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45310 is the best canonical issue because it is the earlier report of the exact same Qwen3.5 Moe `from_pretrained` failure as #45478.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45310 is the best overall issue representative here; it is the only clear duplicate target in the set and appears to be the primary report.", + "best_issue_reason": "#45310 is the strongest issue to keep as the representative duplicate target: same title/root problem as #45478, earlier creation, and more suitable as the primary tracker.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45200", "right": "issue:45588", "accept": false, - "reason": "Different models and different bugs: token-type defaults vs flash-attention `s_aux=None` crash." + "reason": "Different bugs: Gemma 4 text-only finetuning `mm_token_type_ids` defaulting vs a Flash Attention crash on `s_aux=None`." }, { "left": "issue:37428", "right": "issue:44351", "accept": false, - "reason": "Both are import errors, but they involve different symbols and code paths." + "reason": "Both are import/load errors, but they involve different missing symbols and different code paths." }, { "left": "issue:44206", "right": "issue:45823", "accept": false, - "reason": "Unrelated failures in different models/components; not the same bug." + "reason": "Both are device/runtime mismatches, but for different Gemma/Qwen model paths and different failure mechanisms." }, { "left": "issue:29942", "right": "issue:32090", "accept": false, - "reason": "Flash Attention 2 test failures and a Trainer broadcast `NoneType` error are different problems." + "reason": "Flash Attention 2 test failures and `_gpu_broadcast_one` NoneType broadcast errors are unrelated." }, { "left": "issue:43611", "right": "issue:43931", "accept": false, - "reason": "Model loading failures, but with different root causes and different model families." + "reason": "Both are model loading issues, but one is about `base_model_prefix` handling and the other about weight shape mismatch." }, { "left": "issue:43335", "right": "issue:43653", "accept": false, - "reason": "SwitchTransformers config bug vs BigBirdTokenizer special-token registration issue." + "reason": "Different models and failures: SwitchTransformers config vs BigBird tokenizer special-token registration." }, { "left": "issue:43550", "right": "issue:44112", "accept": false, - "reason": "Different subsystems: torch.compile/SDPA vs a stale CI device-override test." + "reason": "Both mention CI/device issues, but they affect different models and do not share a concrete root cause." }, { "left": "issue:43819", "right": "issue:44493", "accept": false, - "reason": "Different code paths: DAC latent reconstruction vs position-id key warnings." - }, - { - "left": "issue:44360", - "right": "issue:44561", - "accept": false, - "reason": "DSA indexer/ReLU discussion is unrelated to the v5 `is_torch_fx_available` removal breakage." + "reason": "Different problems in DAC vs positional-key warnings; same broad audio/model area only." }, { "left": "issue:44855", "right": "issue:45083", "accept": false, - "reason": "Python 3.13 import/parsing issue vs a helper-function behavior bug in qwen3_omni_moe." + "reason": "Python 3.13 import parsing error vs a helper-function behavior bug in qwen3_omni_moe." }, { - "left": "issue:39692", - "right": "issue:43526", + "left": "issue:44360", + "right": "issue:44561", "accept": false, - "reason": "Documentation example errors and a BeitImageProcessorFast label-processing bug are not the same issue." + "reason": "DSA ReLU discussion is unrelated to the v5 removal of `is_torch_fx_available` breaking `trust_remote_code` models." }, { "left": "issue:43335", "right": "issue:43504", "accept": false, - "reason": "Different BEiT/SwitchTransformers loading/configuration bugs." + "reason": "SwitchTransformers config bug and BEiT semantic segmentation legacy-field load failure are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Docs/example errors in SigLIP2 vs a BeitImageProcessorFast `reduce_labels` bug." }, { "left": "issue:45310", "right": "issue:45478", "accept": true, - "reason": "Same exact Qwen3.5 MoE `from_pretrained` regression; identical bug and plausibly one fix." + "reason": "Exact same title and same Qwen3.5 Moe `from_pretrained` regression; #45478 is a duplicate of #45310." }, { "left": "issue:45588", "right": "issue:45859", "accept": false, - "reason": "Different flash-attention-related failures: sinkless-model `s_aux=None` crash vs CPU-offload incompatibility." + "reason": "Different Qwen3.5-related failures: Flash Attention `s_aux=None` crash vs CPU offload incompatibility in rotary embedding." }, { "left": "issue:43335", "right": "issue:45083", "accept": false, - "reason": "Different models and failure modes; not the same underlying defect." + "reason": "Different subsystems and symptoms: SwitchTransformers config vs qwen3_omni_moe feature-length helper behavior." }, { "left": "issue:29127", "right": "issue:32090", "accept": false, - "reason": "LayoutLMv3 error-message clarity issue vs Trainer broadcast type error." + "reason": "LayoutLMv3 error messaging and Trainer `_gpu_broadcast_one` NoneType failures are not the same bug." }, { "left": "issue:39692", "right": "issue:44493", "accept": false, - "reason": "Documentation/example problems do not match the position-id key warning bug." + "reason": "SigLIP2 docs/example issues do not match the positional-id key warnings." }, { "left": "issue:44292", "right": "issue:45520", "accept": false, - "reason": "Different model/runtime problems: Qwen-3-8B-NVFP4 runtime error vs Python 3.13 `flash_attn` import key error." + "reason": "Qwen-3-8B-NVFP4 runtime failure and `flash_attn` import KeyError are different problems." }, { "left": "issue:43720", "right": "issue:44112", "accept": false, - "reason": "Packed-weights loading bug and GraniteSpeech CI stale-device test are unrelated." + "reason": "BitNet packed-weight loading and GraniteSpeech stale device override are unrelated CI/runtime issues." }, { "left": "issue:44448", "right": "issue:44779", "accept": false, - "reason": "Pegasus v4/v5 output mismatch and Deepseek tokenizer regression are different bugs." + "reason": "Pegasus v4/v5 output differences and DeepSeek tokenizer regressions are different code paths." }, { "left": "issue:43335", - "right": "issue:45072", + "right": "issue:44112", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug is unrelated to dtype mismatches in inference." + "reason": "SwitchTransformers config bug and GraniteSpeech CI failure do not share the same underlying defect." }, { "left": "issue:43335", - "right": "issue:44112", + "right": "issue:45072", "accept": false, - "reason": "Different subsystems and no shared code-path problem." + "reason": "SwitchTransformers config and dtype mismatches in bfloat16 inference are unrelated." }, { "left": "issue:38175", "right": "issue:44493", "accept": false, - "reason": "SigLIP2 zero-probability issue is not the same as the unexpected position-id key warning." + "reason": "Unexpected zero probabilities in SigLIP2 and position-id key warnings are not the same bug." }, { "left": "issue:44977", "right": "issue:45229", "accept": false, - "reason": "Flash-attention generation issue vs Gemma4 multi-GPU OOM; different root causes." + "reason": "Flash-attention generation bug for Qwen3.5 is unrelated to Gemma4 multi-GPU CUDA OOM." }, { "left": "issue:42490", "right": "issue:43854", "accept": false, - "reason": "Local-path import warning and GLM-4.7-Flash model-loading test failure are unrelated." + "reason": "Local-path slash import warning and GLM-4.7-Flash test loading failure are different issues." }, { "left": "issue:43606", "right": "issue:44857", "accept": false, - "reason": "CPU offload device mismatch and float16 AMP crash are different failures." + "reason": "CPU offload device mismatch in Bark-small and float16 AMP crash in LwDetrImageLoss are unrelated." }, { "left": "issue:34689", "right": "issue:35141", "accept": false, - "reason": "Both are model-loading regressions, but they involve different models and code paths." + "reason": "Llama 3.2 Vision model-loading breakage and embedding reinit on resize are different root causes." }, { - "left": "issue:40444", - "right": "issue:43526", + "left": "issue:44857", + "right": "issue:45072", "accept": false, - "reason": "IterableDataset multi-image finetuning failure is unrelated to BeitImageProcessorFast label reduction." + "reason": "AMP/CUDA loss crash and bfloat16 dtype mismatch are separate numeric-stability issues." }, { - "left": "issue:44857", - "right": "issue:45072", + "left": "issue:40444", + "right": "issue:43526", "accept": false, - "reason": "Float16 AMP crash and bfloat16 dtype mismatch are different bugs." + "reason": "IterableDataset multi-image finetuning failure and BeitImageProcessorFast `reduce_labels` bug are unrelated." }, { "left": "issue:44561", "right": "issue:44779", "accept": false, - "reason": "v5 trust_remote_code breakage is unrelated to Deepseek tokenizer behavior." + "reason": "Removal of `is_torch_fx_available` breaking remote-code models is unrelated to DeepSeek tokenizer output regression." }, { - "left": "issue:44462", - "right": "issue:44561", + "left": "issue:44448", + "right": "issue:44977", "accept": false, - "reason": "Tokenizer JSON precedence bug and `is_torch_fx_available` removal breakage are different issues." + "reason": "Different model families and symptoms: Pegasus output divergence vs Qwen3.5 flash-attention generation failure." }, { - "left": "issue:44448", - "right": "issue:44977", + "left": "issue:44462", + "right": "issue:44561", "accept": false, - "reason": "Different model regressions with different symptoms and fixes." + "reason": "AutoTokenizer ignoring `tokenizer.json` and v5 removal of `is_torch_fx_available` are unrelated loader regressions." }, { "left": "issue:43723", "right": "issue:44554", "accept": false, - "reason": "AutoTokenizer loading issue vs MPS attention correctness issue; not the same bug." + "reason": "Tokenizer loading in v5 and an MPS attention correctness issue are different bugs." }, { "left": "issue:28218", "right": "issue:32090", "accept": false, - "reason": "Tokenizer spacing bug and Trainer `_gpu_broadcast_one` type error are unrelated." + "reason": "Tokenizer adds an extra space after added tokens; unrelated to Trainer broadcast errors." }, { "left": "issue:43701", "right": "issue:45083", "accept": false, - "reason": "Checkpoint key mismatch and qwen3_omni_moe helper behavior are different problems." - }, - { - "left": "issue:43550", - "right": "issue:45072", - "accept": false, - "reason": "Different compile/inference bugs in different models." + "reason": "Checkpoint key mismatch and qwen3_omni_moe feature-length behavior are different issues." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet: 45310 and 45478 have the same title and 45310 is earlier, so treating 45310 as the canonical issue is reasonable. The soft-edge verdicts are conservative and the rejected pairs are well supported by distinct titles/code paths. No PR-mergeability issue applies here since this is an issue cluster, not a PR pair." + "feedback": "The duplicate target choice is grounded: #45310 and #45478 have identical titles and the earlier report is a sensible canonical issue. The summary stays conservative by describing the rest as broad-theme overlaps rather than asserting they are duplicates. The soft-edge rejections are generally title-grounded and do not overstate evidence. No PR-pair mergeability concerns apply here." }, "evaluator_used": true, "retried": false, @@ -8196,8 +8212,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -8216,6 +8232,7 @@ "issue:43329", "issue:43381", "issue:43412", + "issue:43550", "issue:43575", "issue:43653", "issue:43873", @@ -8225,7 +8242,6 @@ "issue:44186", "issue:44206", "issue:44448", - "issue:44451", "issue:44466", "issue:44479", "issue:44568", @@ -8233,6 +8249,7 @@ "issue:44857", "issue:44977", "issue:44987", + "issue:45083", "issue:45198", "issue:45229", "issue:45245", @@ -8242,7 +8259,6 @@ "issue:45440", "issue:45459", "issue:45491", - "issue:45520", "issue:45563", "issue:45588", "issue:45698", @@ -8252,244 +8268,244 @@ ], "soft_pairs": [ "issue:37428|issue:44987", + "issue:43550|issue:45083", "issue:43927|issue:45698", "issue:32090|issue:33357", - "issue:43873|issue:45229", "issue:29942|issue:35141", "issue:44568|issue:45491", - "issue:41720|issue:45229", + "issue:43873|issue:45229", "issue:44977|issue:45563", + "issue:41720|issue:45229", "issue:45372|issue:45588", "issue:43412|issue:45588", - "issue:45198|issue:45803", "issue:43278|issue:43329", + "issue:45198|issue:45803", "issue:10000|issue:15354", - "issue:43976|issue:45440", "issue:42490|issue:43575", + "issue:43976|issue:45440", "issue:45459|issue:45712", "issue:45372|issue:45406", "issue:44857|issue:45588", "issue:45245|issue:45563", "issue:44448|issue:44466", "issue:45588|issue:45823", - "issue:43653|issue:44743", "issue:43278|issue:43381", + "issue:43653|issue:44743", "issue:43065|issue:43994", "issue:45372|issue:45459", "issue:44206|issue:44479", "issue:43065|issue:45412", "issue:25251|issue:32090", - "issue:44186|issue:45198", - "issue:44451|issue:45520" + "issue:44186|issue:45198" ], - "prepared_review_unit_hash": "c057beb6c7acb903f1225152ae2019f93a6aa246d2993566d2ac897ca5b970ec", + "prepared_review_unit_hash": "881fb2e920dc8370de6271e9cc04bbce392a311f16ccb295a303bf3542b223a6", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36944, - "estimated_input_tokens": 9236, - "estimated_eval_tokens": 18728 + "serialized_chars": 36994, + "estimated_input_tokens": 9249, + "estimated_eval_tokens": 18754 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is highly heterogeneous: most items are distinct bug reports across unrelated subsystems, so there is no strong true canonical. I picked the broadest/open issue as the closest representative, but I would not treat this as a meaningful duplicate cluster.", - "confidence": 0.17, - "canonical_issue_reason": "Issue 43873 is the broadest, open report in the set and has multiple inbound references, so it is the best available representative for an otherwise mixed cluster.", + "summary": "This cluster is a loose collection of unrelated bugs. The soft pairs mostly share only broad themes like model loading, generation, or save/load behavior, but they do not look like the same concrete underlying issue, so none should be merged as duplicates.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45588 is the most central concrete runtime bug in the cluster, and it participates in multiple soft links, making it the best representative issue here.", "canonical_pr_reason": null, - "best_issue_reason": "43873 is the most suitable single issue to anchor the cluster because its scope is broad and it appears to be the only clearly active/open umbrella-style report.", + "best_issue_reason": "issue:45588 is the strongest issue anchor: it is a specific reproducible crash, has the most cluster connectivity, and is more representative than the more niche environment/model-specific reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37428", "right": "issue:44987", "accept": false, - "reason": "Both mention loading/import failures, but they involve different components and failure modes: a flash-attention utility import error vs a model load regression." + "reason": "Different failures: a missing internal flash-attention import vs a model-loading regression for a specific repo." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "Different models and code paths: torch.compile/SDPA on Bamba vs a Qwen3 Omni feature-length helper bug." }, { "left": "issue:43927", "right": "issue:45698", "accept": false, - "reason": "Both concern save/load behavior, but one is about DiaConfig losing custom token IDs and the other about loading the wrong custom module after save_pretrained; different bugs." + "reason": "Both involve save/load, but one loses custom token IDs in config while the other resolves the wrong custom module." }, { "left": "issue:32090", "right": "issue:33357", "accept": false, - "reason": "Different problems: Trainer distributed broadcast NoneType error vs MacOS bus error with a community CLIP model." - }, - { - "left": "issue:43873", - "right": "issue:45229", - "accept": false, - "reason": "One is a general offloading/quantization report; the other is a multi-GPU CUDA OOM for a specific Gemma model. Not the same underlying bug." + "reason": "Different subsystems and symptoms: Trainer broadcast NoneType vs a MacOS bus error loading a CLIP model." }, { "left": "issue:29942", "right": "issue:35141", "accept": false, - "reason": "Flash Attention 2 test failures and token embedding reinit after resize are unrelated code-paths." + "reason": "Unrelated bugs: Flash Attention 2 test failures vs token embedding resize reinitializing the output embedding." }, { "left": "issue:44568", "right": "issue:45491", "accept": false, - "reason": "Tokenizer special-token behavior and Gemma3 NaN embeddings on mixed-length GPU batches are different defects." + "reason": "Tokenizer special-token handling is unrelated to Gemma3 NaN embeddings from sliding-window attention." }, { - "left": "issue:41720", + "left": "issue:43873", "right": "issue:45229", "accept": false, - "reason": "Auto device-mapping cudaErrorAssert on A800 is not the same as a multi-GPU OOM during Gemma4 inference." + "reason": "Both are memory-related at a high level, but one is quantization/offloading behavior and the other is a multi-GPU OOM." }, { "left": "issue:44977", "right": "issue:45563", "accept": false, - "reason": "Qwen3.5 flash-attention generation regression and a stale paged-generate warning are unrelated." + "reason": "Different issues: a flash-attention generation regression vs a stale warning in paged generate()." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Different concrete bugs: Qwen3 auto device mapping hits cudaErrorAssert, while Gemma4 31B hits an OOM." }, { "left": "issue:45372", "right": "issue:45588", "accept": false, - "reason": "Gemma 4 processor loading import failure and a flash_attention integration crash on sink-less models are different failure paths." + "reason": "Different failure points: Gemma4 processor import breaks on ReasoningEffort, while flash_attention.py crashes on s_aux=None." }, { "left": "issue:43412", "right": "issue:45588", "accept": false, - "reason": "Executorch export issues for Gemma3n and a flash_attention runtime AttributeError are unrelated." + "reason": "Export-time executorch/erfinv issues are unrelated to a runtime flash-attention AttributeError." }, { - "left": "issue:45198", - "right": "issue:45803", + "left": "issue:43278", + "right": "issue:43329", "accept": false, - "reason": "Wav2Vec2 save/tokenization failure and a bare-except in Fuyu tensor conversion are different bugs." + "reason": "Embedding dtype drift in eval is unrelated to undefined video-branch helpers in multimodal token counting." }, { - "left": "issue:43278", - "right": "issue:43329", + "left": "issue:45198", + "right": "issue:45803", "accept": false, - "reason": "Embedding dtype drift on evaluate and a multimodal-token counting bug in video branch are unrelated." + "reason": "Different code paths: wav2vec2 save/tokenization failure vs a bare-except bug in Fuyu tensor conversion." }, { "left": "issue:10000", "right": "issue:15354", "accept": false, - "reason": "German DistilBertModel issue and GeneratorExp/JIT script export issue have no shared code-path." + "reason": "Completely different models and errors: DistilBert issue vs torch.jit GeneratorExp export limitation." }, { - "left": "issue:43976", - "right": "issue:45440", + "left": "issue:42490", + "right": "issue:43575", "accept": false, - "reason": "Python version compatibility for Transformers 5.1.0 and DeepseekV3MoE divergence are distinct problems." + "reason": "A trailing-slash import warning is unrelated to a tensor-parallel OOM on Qwen2-57B." }, { - "left": "issue:42490", - "right": "issue:43575", + "left": "issue:43976", + "right": "issue:45440", "accept": false, - "reason": "Trailing-slash local path import warning and Qwen2-57B tensor-parallel OOM are unrelated." + "reason": "Environment compatibility and model-implementation divergence are different bugs." }, { "left": "issue:45459", "right": "issue:45712", "accept": false, - "reason": "Tokenizer protobuf error masking and leftover dummy classes leaking into dir(transformers) are separate maintenance issues." + "reason": "Both touch import fallbacks, but one hides tokenizer errors while the other is leftover dummy-class cleanup." }, { "left": "issue:45372", "right": "issue:45406", "accept": false, - "reason": "Gemma 4 processor import problem and Gemma4Processor missing _tokenizer in serve are related by model family but not the same concrete bug." + "reason": "Same model family, but distinct bugs: missing ReasoningEffort import vs Gemma4Processor lacking _tokenizer in serve." }, { "left": "issue:44857", "right": "issue:45588", "accept": false, - "reason": "AMP/CUDA crash in LwDetrImageLoss and flash_attention AttributeError on sink-less models are different subsystems and failure modes." + "reason": "Different components: LwDetrImageLoss AMP crash vs flash_attention s_aux=None AttributeError." }, { "left": "issue:45245", "right": "issue:45563", "accept": false, - "reason": "Category-count runtime error and paged generate warning are not the same bug." + "reason": "No overlap beyond being generation-related; category limit error and stale warning are unrelated." }, { "left": "issue:44448", "right": "issue:44466", "accept": false, - "reason": "Different-output regression for Pegasus vs inconsistent lm_head serialization tied to device are separate issues." + "reason": "Both are v5 regressions, but one is output divergence and the other is tied-weight serialization behavior." }, { "left": "issue:45588", "right": "issue:45823", "accept": false, - "reason": "Flash attention sink-less crash and Gemma4 PLE device mismatch during forward are different code paths." + "reason": "Gemma4 flash-attention crash and Gemma4 PLE device mismatch are separate forward-path bugs." }, { - "left": "issue:43653", - "right": "issue:44743", + "left": "issue:43278", + "right": "issue:43381", "accept": false, - "reason": "BigBirdTokenizer special-token registration and Qwen3_5 recurrent-state reset are unrelated." + "reason": "Eval-mode gradient-checkpointing restriction is unrelated to the BF16-to-FP32 embedding dtype change." }, { - "left": "issue:43278", - "right": "issue:43381", + "left": "issue:43653", + "right": "issue:44743", "accept": false, - "reason": "Embedding dtype regression and gradient checkpointing being disallowed in eval mode are distinct issues." + "reason": "Tokenizer special-token registration and recurrent-state reset are unrelated bugs." }, { "left": "issue:43065", "right": "issue:43994", "accept": false, - "reason": "Dummy Conv2d in Sam3PixelDecoder and SigLIP2 nonsense outputs are unrelated." + "reason": "A dummy Conv2d placeholder in Sam3PixelDecoder is not the same as Siglip2 producing bad outputs." }, { "left": "issue:45372", "right": "issue:45459", "accept": false, - "reason": "Gemma 4 processor import breakage and protobuf decode error masking in tokenizers are different bugs." + "reason": "Distinct import-path bugs: missing ReasoningEffort vs tokenizer errors being swallowed by a broad exception." }, { "left": "issue:44206", "right": "issue:44479", "accept": false, - "reason": "LasrFeatureExtractor center-arg regression and Qwen video-input regression are unrelated." + "reason": "Different regressions: unsupported center arg in LasrFeatureExtractor vs video-input failure in Qwen VL models." }, { "left": "issue:43065", "right": "issue:45412", "accept": false, - "reason": "Dummy Conv2d placeholder issue and RT-DETR memory release problem are separate code paths." + "reason": "A dummy layer placeholder and a model memory-leak issue are not the same underlying bug." }, { "left": "issue:25251", "right": "issue:32090", "accept": false, - "reason": "Pipeline top_k nesting change and Trainer _gpu_broadcast_one NoneType error are unrelated." + "reason": "Pipeline output-shape change is unrelated to Trainer broadcast failures." }, { "left": "issue:44186", "right": "issue:45198", "accept": false, - "reason": "LayoutLMv2Tokenizer NER/padding crash and Wav2Vec2 save/tokenization failure are different tokenizer bugs." - }, - { - "left": "issue:44451", - "right": "issue:45520", - "accept": false, - "reason": "Model loading failure for ScandiBERT and Python 3.13 flash_attn import KeyError are unrelated." + "reason": "Different tokenizers and symptoms: LayoutLMv2 NER/padding crashes vs wav2vec2 save/tokenization failure." } ] }, @@ -8508,8 +8524,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -8528,11 +8544,9 @@ "issue:43066", "issue:43257", "issue:43278", - "issue:43334", "issue:43381", "issue:43421", "issue:43526", - "issue:43531", "issue:43611", "issue:43653", "issue:43701", @@ -8546,6 +8560,7 @@ "issue:44279", "issue:44361", "issue:44442", + "issue:44451", "issue:44462", "issue:44610", "issue:44871", @@ -8556,17 +8571,19 @@ "issue:45412", "issue:45440", "issue:45491", + "issue:45520", "issue:45593", "issue:45834", "issue:45859", "issue:46121" ], "soft_pairs": [ + "issue:44451|issue:45520", "issue:43066|issue:44462", "issue:44871|issue:44977", - "issue:42915|issue:42994", "issue:44361|issue:45198", "issue:43611|issue:46121", + "issue:42915|issue:42994", "issue:43994|issue:45412", "issue:43756|issue:45440", "issue:33357|issue:39692", @@ -8584,202 +8601,201 @@ "issue:43381|issue:43701", "issue:43526|issue:45834", "issue:43526|issue:44186", - "issue:43761|issue:45859", "issue:29127|issue:33357", + "issue:43761|issue:45859", "issue:29942|issue:36010", - "issue:45103|issue:45372", - "issue:43334|issue:43531" + "issue:45103|issue:45372" ], - "prepared_review_unit_hash": "560fce233a4f8614d90b8d896e76d37fc090f793688790308a2fa427f3ac16c0", + "prepared_review_unit_hash": "c46dbc793e9575e2c45d5f6bf554438193d7249926ae4517512a7e52da60c046", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 27, - "serialized_chars": 35192, - "estimated_input_tokens": 8798, - "estimated_eval_tokens": 17852 + "serialized_chars": 35145, + "estimated_input_tokens": 8787, + "estimated_eval_tokens": 17830 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "All proposed soft links look like false positives: they share broad language (tokenizer, quantization, trainer, model loading) but point to different concrete bugs, model families, or code paths.", - "confidence": 0.94, - "canonical_issue_reason": "No single canonical issue stands out because the set is heterogeneous and none of the soft pairs describe the same underlying defect.", + "summary": "The cluster is heterogeneous: the soft pairs mostly share broad Transformer-area vocabulary (tokenizers, loading, quantization, vision, trainer) but not the same concrete bug or code path. I would not merge any of the soft pairs as duplicates. No PRs are present.", + "confidence": 0.81, + "canonical_issue_reason": "No true canonical duplicate stands out; if one must be chosen, issue 43873 is the broadest open runtime/loading/quantization/offloading report with the most cluster-like scope, but it does not actually subsume the others.", "canonical_pr_reason": null, - "best_issue_reason": "No issue is a strong global representative for duplicate consolidation here; the closest matches are still materially different bugs.", + "best_issue_reason": "Issue 43873 is the best representative only in the sense of being the broadest, active infra-style bug in the set; however, the surrounding issues are too diverse for a strong canonical fit.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:44451", + "right": "issue:45520", + "accept": false, + "reason": "Both are import/load failures, but one is model loading and the other is a Python 3.13 flash_attn import key error; different causes and fixes." + }, { "left": "issue:43066", "right": "issue:44462", "accept": false, - "reason": "Different failure modes: tokenizer decoder-type regression vs AutoTokenizer ignoring tokenizer.json." + "reason": "Both concern tokenizers, but one is an eos/decoder-type mismatch and the other is repository tokenizer.json selection; not the same bug." }, { "left": "issue:44871", "right": "issue:44977", "accept": false, - "reason": "Different models and symptoms; one is eos_token_id config inconsistency, the other is flash-attention generation breakage." - }, - { - "left": "issue:42915", - "right": "issue:42994", - "accept": false, - "reason": "Both involve quantization, but one is an FP8 config failure and the other is a model-saving failure." + "reason": "Different models and failure modes: Gemma EOS config mismatch vs Qwen flash-attention generation issues." }, { "left": "issue:44361", "right": "issue:45198", "accept": false, - "reason": "Different tokenizers and code paths; not the same bug." + "reason": "Unrelated tokenizer/task failures affecting different model families and code paths." }, { "left": "issue:43611", "right": "issue:46121", "accept": false, - "reason": "One is a base_model_prefix loading regression; the other is a rope-validation type error." + "reason": "One is a model-loading base_model_prefix regression; the other is a rope-validation helper TypeError. No shared concrete defect." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Both touch quantization-adjacent behavior, but one is MoE weight conversion under accelerate/deepspeed and the other is quantized model saving." }, { "left": "issue:43994", "right": "issue:45412", "accept": false, - "reason": "Unrelated bugs: bad SigLIP2 outputs vs RT-DETR memory not being released." + "reason": "Nonsensical outputs in SigLIP2 vs model memory not being released in RT-DETR are unrelated bugs." }, { "left": "issue:43756", "right": "issue:45440", "accept": false, - "reason": "Both concern model parity with another implementation, but they affect different architectures and underlying code paths." + "reason": "Different model implementations and different structural divergences; not the same underlying issue." }, { "left": "issue:33357", "right": "issue:39692", "accept": false, - "reason": "Different issues: MacOS bus error on CLIP vs documentation/example errors for SigLIP2." + "reason": "Both mention CLIP/SigLIP-family models, but one is a Mac bus error and the other is documentation/example mistakes." }, { "left": "issue:29127", "right": "issue:34567", "accept": false, - "reason": "Unrelated trainer/error-message issue vs TrainerState token-count update bug." + "reason": "Different subsystems: LayoutLMv3 error messaging vs TrainerState token counting." }, { "left": "issue:32090", "right": "issue:33290", "accept": false, - "reason": "Different training failures: NoneType broadcast vs OOM with Adafactor/DeepSpeed." + "reason": "Both involve Trainer/DeepSpeed contexts, but one is a NoneType broadcast error and the other is an optimizer OOM." }, { "left": "issue:42490", "right": "issue:43873", "accept": false, - "reason": "Local-path import warning is unrelated to quantization offloading behavior." + "reason": "Local-path import warning and quantization/offloading behavior are distinct loading-path issues." }, { "left": "issue:29942", "right": "issue:30064", "accept": false, - "reason": "Flash Attention 2 test failures and void segmentation-map processing are separate bugs." + "reason": "Flash Attention test failures and void segmentation map processing are unrelated." }, { "left": "issue:43257", "right": "issue:43278", "accept": false, - "reason": "Different symptoms and subsystems: MOE weight conversion vs dtype drift during evaluation." + "reason": "Qwen3 MoE weight conversion under accelerate/deepspeed is not the same as BF16-to-FP32 dtype drift during eval." }, { "left": "issue:43421", "right": "issue:45491", "accept": false, - "reason": "Runtime special-token post-processing bug is unrelated to Gemma3 NaN embeddings from sliding-window padding behavior." + "reason": "Runtime special-token post-processing and Gemma3 NaN embeddings from sliding-window attention are different defects." }, { "left": "issue:44265", "right": "issue:45198", "accept": false, - "reason": "torch.export/torch_compilable_check failure is unrelated to Wav2Vec2 save/tokenization breakage." + "reason": "torch.export/torch_compilable_check failure is unrelated to Wav2Vec2 save_pretrained/tokenization problems." }, { "left": "issue:44610", "right": "issue:45412", "accept": false, - "reason": "Processor image-size mismatch and a memory leak are distinct problems." + "reason": "Processor input-size mismatch and RT-DETR memory reclamation are different bugs." }, { "left": "issue:45440", "right": "issue:45593", "accept": false, - "reason": "Both are model-specific correctness issues, but they affect different models and implementation details." + "reason": "DeepseekV3MoE implementation drift and D-FINE auxiliary-loss behavior are not the same code-path problem." }, { "left": "issue:43653", "right": "issue:44442", "accept": false, - "reason": "Tokenizer special-token registration bug is not the same as AutoTokenizer failing to load a tokenizer class." + "reason": "Both involve tokenizers, but one is BigBird special-token registration/empty decode and the other is AutoTokenizer loading a specific tokenizer class." }, { "left": "issue:43824", "right": "issue:44279", "accept": false, - "reason": "Specific missing-import regression vs a generic dependency issue; not the same root cause." + "reason": "An import error for a missing model class is not the same as a generic dependency issue." }, { "left": "issue:43381", "right": "issue:43701", "accept": false, - "reason": "Gradient checkpointing in eval mode and checkpoint resume key mismatch are unrelated." + "reason": "Gradient checkpointing in eval mode and resume_from_checkpoint key mismatch are distinct trainer-state issues." }, { "left": "issue:43526", "right": "issue:45834", "accept": false, - "reason": "Different vision/OCR issues: label reduction bug vs long-input index error." + "reason": "BeitImageProcessorFast label reduction and Kosmos2.5 long OCR indexing are different image-processing bugs." }, { "left": "issue:43526", "right": "issue:44186", "accept": false, - "reason": "Different tokenization/processor bugs with no shared code-path evidence." + "reason": "Different tokenizer/image-processor paths and unrelated failure modes." }, { - "left": "issue:43761", - "right": "issue:45859", + "left": "issue:29127", + "right": "issue:33357", "accept": false, - "reason": "CLIPVision hidden_states regression is unrelated to Qwen3 MoE CPU-offload incompatibility." + "reason": "LayoutLMv3 error-message clarity and a CLIP bus error on macOS are unrelated." }, { - "left": "issue:29127", - "right": "issue:33357", + "left": "issue:43761", + "right": "issue:45859", "accept": false, - "reason": "LayoutLMv3 error-message clarity and CLIP MacOS bus error are unrelated." + "reason": "CLIPVisionModel hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are separate model-specific bugs." }, { "left": "issue:29942", "right": "issue:36010", "accept": false, - "reason": "Flash Attention 2 test failures do not match the GenerationMixin import error." + "reason": "Flash Attention test failures do not match a GenerationMixin import regression." }, { "left": "issue:45103", "right": "issue:45372", "accept": false, - "reason": "Auto-docstring annotation handling and Gemma 4 processor import failure are unrelated." - }, - { - "left": "issue:43334", - "right": "issue:43531", - "accept": false, - "reason": "Both involve Qwen3 family models, but one is a missing pad_token_id load error and the other is a sliding_window behavior issue." + "reason": "Auto-docstring crashes under future annotations and Gemma 4 processor loading failing on a missing mistral_common symbol are unrelated." } ] }, @@ -8798,32 +8814,32 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ "issue:29127", "issue:33290", "issue:34567", - "issue:34689", "issue:35141", "issue:42490", "issue:42915", "issue:43232", "issue:43262", "issue:43316", + "issue:43334", "issue:43381", "issue:43388", "issue:43441", "issue:43525", "issue:43526", + "issue:43531", "issue:43575", "issue:43646", "issue:43701", "issue:43723", "issue:43761", - "issue:43873", "issue:43931", "issue:44062", "issue:44077", @@ -8853,14 +8869,15 @@ "issue:45907" ], "soft_pairs": [ + "issue:43334|issue:43531", "issue:44625|issue:45584", "issue:29127|issue:35141", "issue:44977|issue:45245", "issue:42915|issue:43931", "issue:44442|issue:45072", "issue:43526|issue:44857", - "issue:43262|issue:44206", "issue:44464|issue:45491", + "issue:43262|issue:44206", "issue:42490|issue:43381", "issue:43575|issue:43701", "issue:43575|issue:44387", @@ -8878,206 +8895,198 @@ "issue:43761|issue:44188", "issue:44442|issue:44898", "issue:42490|issue:44743", - "issue:33290|issue:34567", - "issue:33290|issue:34689", - "issue:43381|issue:43873" + "issue:33290|issue:34567" ], - "prepared_review_unit_hash": "87e61702da4ba12acc1196f1854c47abf19a93d1a005ddbff32da5f196399f59", + "prepared_review_unit_hash": "64428ce8bd1ca603d69cd72db9eeeccefa9462b340a01c766ff6585060cd8629", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 28, - "serialized_chars": 36176, - "estimated_input_tokens": 9044, - "estimated_eval_tokens": 18344 + "soft_pair_count": 27, + "serialized_chars": 35905, + "estimated_input_tokens": 8977, + "estimated_eval_tokens": 18210 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated issue reports; the soft-similarity links are lexical false positives, so none should be merged. No clear duplicate cluster emerged.", - "confidence": 0.94, - "canonical_issue_reason": "No true canonical issue stands out because the items span unrelated bugs and regressions. If forced to pick a representative, #43873 has the broadest discussion/inbound references, but it does not subsume the rest.", + "summary": "No duplicate cluster here: the items are a set of unrelated bug reports spanning configs, tokenizers, generation, vision/audio processors, attention kernels, and memory/runtime issues. All soft pairs should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue; the items are unrelated regressions across different subsystems, so none is a safe duplicate anchor.", "canonical_pr_reason": null, - "best_issue_reason": "#43873 is the most central by activity and inbound references, but only as a loose representative; it is not a genuine umbrella duplicate for this set.", + "best_issue_reason": "No strong cluster representative for the same reason: the reports do not describe one underlying bug or change.", "best_pr_reason": null, "soft_edge_verdicts": [ + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Different bugs: Qwen3-VL pad_token_id AttributeError vs Qwen3-MoE sliding_window behavior." + }, { "left": "issue:44625", "right": "issue:45584", "accept": false, - "reason": "Different features and failures: Qwen3.5 config propagation vs Whisper generation on empty transcription." + "reason": "Unrelated: Qwen3.5 num_labels propagation vs Whisper empty-transcription/chat-template failure." }, { "left": "issue:29127", "right": "issue:35141", "accept": false, - "reason": "Unrelated bugs: LayoutLMv3 error messaging vs token-embedding resizing and post-init reinitialization." + "reason": "LayoutLMv3 error messaging is unrelated to token embedding resize/post_init reinitialization." }, { "left": "issue:44977", "right": "issue:45245", "accept": false, - "reason": "Both mention model/runtime failures, but one is a flash-attention generation bug and the other is a category-cardinality error." + "reason": "Different failure modes: Qwen3.5 flash-attention generation vs category-count runtime limit." }, { "left": "issue:42915", "right": "issue:43931", "accept": false, - "reason": "Same model family, but different underlying problems: FineGrainedFP8Config failure vs Qwen3-VL weight-shape mismatch." + "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are separate model-loading bugs." }, { "left": "issue:44442", "right": "issue:45072", "accept": false, - "reason": "Different subsystems and symptoms: tokenizer loading vs dtype mismatch in inference wrappers." + "reason": "Tokenizer loading failure is unrelated to dtype mismatches in inference wrappers." }, { "left": "issue:43526", "right": "issue:44857", "accept": false, - "reason": "Different image-processing and training-time bugs; no shared code-path evidence." + "reason": "BeitImageProcessorFast label reduction bug is unrelated to LwDetrImageLoss AMP/CUDA crash." }, { - "left": "issue:43262", - "right": "issue:44206", + "left": "issue:44464", + "right": "issue:45491", "accept": false, - "reason": "Audio processor chat-template sampling-rate default vs LASR feature-extractor argument crash are distinct issues." + "reason": "Compiled forward chunked-generation inconsistency is not the same as Gemma3 sliding-window NaNs." }, { - "left": "issue:44464", - "right": "issue:45491", + "left": "issue:43262", + "right": "issue:44206", "accept": false, - "reason": "Chunked generation under compiled forward is unrelated to Gemma3 NaN embeddings from sliding-window attention." + "reason": "Audio processor sampling-rate default mismatch vs unsupported center arg in LasrFeatureExtractor." }, { "left": "issue:42490", "right": "issue:43381", "accept": false, - "reason": "Local-path import warning is unrelated to gradient checkpointing eval-mode restrictions." + "reason": "Local-path trailing-slash warning is unrelated to gradient-checkpointing eval-mode behavior." }, { "left": "issue:43575", "right": "issue:43701", "accept": false, - "reason": "OOM on tensor-parallel Qwen loading is unrelated to resume-from-checkpoint key mismatches." + "reason": "TP OOM on Qwen2-57B and resume_from_checkpoint key mismatch are different problems." }, { "left": "issue:43575", "right": "issue:44387", "accept": false, - "reason": "Both are memory-related, but one is tensor-parallel load OOM and the other is int4 quantization reserved-memory growth; different mechanisms." + "reason": "Both mention OOM, but one is tensor parallel loading and the other is int4 reserved-memory regression." }, { "left": "issue:44062", "right": "issue:45907", "accept": false, - "reason": "Tokenizer AddedToken keyword collision is unrelated to list_repo_files forwarding revision." + "reason": "AddedToken kwarg collision is unrelated to list_repo_files not forwarding revision." }, { "left": "issue:44743", "right": "issue:44811", "accept": false, - "reason": "Recurrent-state cache reset bug is unrelated to Whisper batch_decode skip_special_tokens handling." + "reason": "Recurrent-state reset in Qwen3.5 is unrelated to Whisper batch_decode skip_special_tokens handling." }, { "left": "issue:43441", "right": "issue:45588", "accept": false, - "reason": "Both touch flash attention, but they concern different models and different crashes." + "reason": "Ministral-3 FlashAttention failure and sink-less model s_aux=None crash are different code paths." }, { "left": "issue:43646", "right": "issue:44077", "accept": false, - "reason": "General v5 custom-model init breakage is not the same as patchtsmixer's optional post_init policy issue." + "reason": "Broad custom-model-init breakage is not the same as patchtsmixer post_init allowance." }, { "left": "issue:43761", "right": "issue:44079", "accept": false, - "reason": "CLIP hidden_states regression and ModelOutput key assignment are separate code paths." + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment bug are unrelated." }, { "left": "issue:45003", "right": "issue:45440", "accept": false, - "reason": "Unsafe sys.modules access is unrelated to DeepseekV3MoE divergence from remote implementation." + "reason": "Unsafe sys.modules access is unrelated to DeepseekV3MoE implementation divergence." }, { "left": "issue:43232", "right": "issue:43388", "accept": false, - "reason": "Generation kwargs update after sync_gpus is unrelated to gather_for_metrics dropping tuple labels." + "reason": "Generation kwargs update after sync_gpus and metric-gather label truncation are separate bugs." }, { "left": "issue:43723", "right": "issue:45479", "accept": false, - "reason": "Tokenizer loading regression and sequence-classification zero-loss bug are unrelated." + "reason": "Tokenizer loading regression and single-label-classification zero-loss bug do not share a code path." }, { "left": "issue:44479", "right": "issue:45290", "accept": false, - "reason": "Video-input regression for Qwen VL models is unrelated to chat-template crashes with tool-call assistant messages." + "reason": "Video-input regression in Qwen VL models is unrelated to chat-template tokenization with tool calls." }, { "left": "issue:43316", "right": "issue:43525", "accept": false, - "reason": "Both are config API issues, but they affect different models and different missing fields." + "reason": "API discrepancy in Gemma3TextConfig is not the same as missing pad_token_id in Llama4Config." }, { "left": "issue:44442", "right": "issue:45803", "accept": false, - "reason": "Tokenizer loading failure is unrelated to the bare-except tensor conversion bug." + "reason": "AutoTokenizer failure and bare-except in FuyuBatchFeature are unrelated." }, { "left": "issue:43761", "right": "issue:44188", "accept": false, - "reason": "Hidden-states output regression is unrelated to torch.compile attention-kernel branching." + "reason": "CLIPVisionModel hidden_states regression and torch.compile attention-kernel divergence are different issues." }, { "left": "issue:44442", "right": "issue:44898", "accept": false, - "reason": "Tokenizer loading and Perceiver resolution handling are different bugs." + "reason": "Tokenizer loading bug is unrelated to Perceiver non-default-resolution interpolation failure." }, { "left": "issue:42490", "right": "issue:44743", "accept": false, - "reason": "Import-path warning is unrelated to recurrent-state reset in modular_qwen3_5." + "reason": "Import warning on trailing slash path and Qwen3.5 recurrent-state reset are unrelated." }, { "left": "issue:33290", "right": "issue:34567", "accept": false, - "reason": "Deepspeed Adafactor OOM and TrainerState token counter not updating are unrelated." - }, - { - "left": "issue:33290", - "right": "issue:34689", - "accept": false, - "reason": "Both involve model training/loading, but the first is optimizer OOM and the second is a specific Llama 3.2 Vision loading regression." - }, - { - "left": "issue:43381", - "right": "issue:43873", - "accept": false, - "reason": "Gradient-checkpointing eval-mode restriction and quantization offloading problems are separate issues." + "reason": "AdaFactor/DeepSpeed OOM and TrainerState num_input_tokens_seen not updating are unrelated." } ] }, @@ -9096,8 +9105,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -9116,6 +9125,7 @@ "issue:43065", "issue:43295", "issue:43316", + "issue:43381", "issue:43404", "issue:43441", "issue:43493", @@ -9150,8 +9160,10 @@ "issue:45759" ], "soft_pairs": [ - "issue:34689|issue:43611", + "issue:33290|issue:34689", "issue:33290|issue:35141", + "issue:43381|issue:43873", + "issue:34689|issue:43611", "issue:42907|issue:45698", "issue:44077|issue:44479", "issue:43931|issue:45216", @@ -9160,8 +9172,8 @@ "issue:43316|issue:45759", "issue:41628|issue:45372", "issue:34689|issue:36010", - "issue:43749|issue:43931", "issue:33290|issue:36331", + "issue:43749|issue:43931", "issue:43873|issue:44871", "issue:43643|issue:44913", "issue:44190|issue:44496", @@ -9171,196 +9183,208 @@ "issue:43065|issue:44617", "issue:43404|issue:43441", "issue:42175|issue:43316", - "issue:44162|issue:44792", "issue:25251|issue:28282", + "issue:44162|issue:44792", "issue:43295|issue:43825", "issue:43673|issue:44743", "issue:16998|issue:22355" ], - "prepared_review_unit_hash": "ea8b31595b11ea66b563236b43d7d7ff6c29fc9ffaa44a6642fb1bae19f522a3", + "prepared_review_unit_hash": "1ab02f30bf2128d19719b5654b8ccd182439d4679d8b413322fb55027fb29256", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 26, - "serialized_chars": 34950, - "estimated_input_tokens": 8738, - "estimated_eval_tokens": 17732 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36095, + "estimated_input_tokens": 9024, + "estimated_eval_tokens": 18304 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is a heterogeneous mix of unrelated bug reports and regressions. The soft pairs share only broad subsystem keywords (loading, config, processors, etc.) but not the same underlying failure mode or fix, so none should be merged as duplicates.", - "confidence": 0.98, - "canonical_issue_reason": "No canonical issue: the items are not one duplicate cluster and do not share a single root cause.", + "summary": "The soft pairs are mostly superficial text similarities around loading/import/config, but they describe different concrete bugs, models, or code paths. None should be merged as duplicate pairs.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43749 is the best representative issue to anchor triage because it has the broadest loading/infra scope and the strongest discussion/inbound-ref signal in this set.", "canonical_pr_reason": null, - "best_issue_reason": "No single issue is a good representative; the collection spans unrelated problems across loading, config, generation, quantization, and packaging.", + "best_issue_reason": "issue:43749 is the highest-signal, most broadly relevant issue here, but it still does not match the other items closely enough to serve as a true duplicate canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:34689", - "right": "issue:43611", + "left": "issue:33290", + "right": "issue:34689", "accept": false, - "reason": "Both involve model loading regressions, but one is a specific Llama 3.2 Vision break in 4.46.2 and the other is a v5 base_model_prefix loading issue; different code paths." + "reason": "Different failures: Adafactor/DeepSpeed OOM vs a specific Llama 3.2 vision model-loading regression." }, { "left": "issue:33290", "right": "issue:35141", "accept": false, - "reason": "OOM with Adafactor/DeepSpeed is unrelated to output embeddings being reinitialized after resize_token_embeddings." + "reason": "Training OOM and embedding-resize post_init reinitialization are unrelated code paths." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Eval-mode gradient checkpointing issue vs quantization/offloading bug." + }, + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both are loading regressions, but one is a model-specific Llama 3.2 breakage and the other is a base_model_prefix loader regression." }, { "left": "issue:42907", "right": "issue:45698", "accept": false, - "reason": "Dequantized save failures and loading the wrong custom module after save_pretrained are different serialization bugs." + "reason": "Both involve save/load, but one is dequantized model saving and the other is wrong custom module resolution after save_pretrained." }, { "left": "issue:44077", "right": "issue:44479", "accept": false, - "reason": "A patchtsmixer post_init restriction is unrelated to the Qwen video-input regression." + "reason": "Unrelated issues: patchtsmixer post_init validation vs Qwen video-input regression." }, { "left": "issue:43931", "right": "issue:45216", "accept": false, - "reason": "Weight-shape mismatch on load and incorrect checkpoint saving are opposite sides of model I/O and affect different models." + "reason": "Different models and failures: Qwen3-VL weight-shape mismatch vs Qwen3.5 checkpoint saving corruption." }, { "left": "issue:43720", "right": "issue:44442", "accept": false, - "reason": "BitNet packed-weight handling during accelerate loading has nothing to do with AutoTokenizer failing on FastSpeech2ConformerTokenizer." + "reason": "Packed-weight accelerate loading bug vs tokenizer import failure." }, { "left": "issue:43493", "right": "issue:43827", "accept": false, - "reason": "SigLIP2 HF/JAX parity is a model implementation bug; the pipeline() issue is a docs/removal mismatch." + "reason": "SigLIP2 implementation mismatch vs stale docs referencing pipeline() after v5 removals." }, { "left": "issue:43316", "right": "issue:45759", "accept": false, - "reason": "Gemma3TextConfig API inconsistency is not the same as failing to unwrap text_config in composite Qwen configs." + "reason": "Different API mismatches in different model families; not the same bug." }, { "left": "issue:41628", "right": "issue:45372", "accept": false, - "reason": "AutoImageProcessor import failure and mistral_common ReasoningEffort import failure are separate dependency/import problems." + "reason": "Both are import errors, but they involve different modules and different root causes." }, { "left": "issue:34689", "right": "issue:36010", "accept": false, - "reason": "A model-loading regression for a vision model is unrelated to the missing GenerationMixin import path." + "reason": "Separate loader/import regressions with different missing symbols and code paths." }, { - "left": "issue:43749", - "right": "issue:43931", + "left": "issue:33290", + "right": "issue:36331", "accept": false, - "reason": "FSDP CPU RAM efficient loading and Qwen3-VL shape mismatch are both loading-related but not the same bug or fix." + "reason": "Optimizer OOM under DeepSpeed is unrelated to the CustomTrainer compute_loss signature break." }, { - "left": "issue:33290", - "right": "issue:36331", + "left": "issue:43749", + "right": "issue:43931", "accept": false, - "reason": "An optimizer memory issue is unrelated to a trainer compute_loss signature change." + "reason": "Different loading subsystems: FSDP CPU-efficient loading vs a Qwen3-VL shape mismatch." }, { "left": "issue:43873", "right": "issue:44871", "accept": false, - "reason": "Quantization/offloading behavior and eos_token_id config mismatch are different features and failure modes." + "reason": "Quantization/offloading behavior vs eos_token_id configuration inconsistency." }, { "left": "issue:43643", "right": "issue:44913", "accept": false, - "reason": "trust_remote_code dropping config fields is a loading/parsing issue, while rotary_pct resetting is a config round-trip bug." + "reason": "Remote-code config deserialization issue vs GPTNeoX config field reverting on reload." }, { "left": "issue:44190", "right": "issue:44496", "accept": false, - "reason": "Local dataset loading in an example script is unrelated to an unrecognized model config lacking model_type." + "reason": "Dataset loading script problem vs unrecognized model/config loading error." }, { "left": "issue:44855", "right": "issue:45137", "accept": false, - "reason": "Python 3.13 torch.jit parsing failure and DeepSpeed ZeRO3 deque underflow are unrelated runtime errors." + "reason": "Python 3.13 torch.jit parsing problem vs DeepSpeed ZeRO3 deque error." }, { "left": "issue:43540", "right": "issue:44514", "accept": false, - "reason": "Video input processing in Qwen3OmniMoe and batched apply_chat_template crashes in Qwen2_5_VLProcessor are different processor bugs." + "reason": "Different multimodal bugs: video processing failure vs batched chat-template padding crash." }, { "left": "issue:43316", "right": "issue:44877", "accept": false, - "reason": "Gemma3TextConfig API mismatch and strict granite_speech config loading are different config validation problems." + "reason": "Gemma3TextConfig API inconsistency vs strict config rejection for granite_speech." }, { "left": "issue:43065", "right": "issue:44617", "accept": false, - "reason": "A dummy Conv2d in Sam3PixelDecoder is unrelated to Sam3Video CUDA OOM." + "reason": "Dummy Conv2d design issue vs CUDA OOM in Sam3Video." }, { "left": "issue:43404", "right": "issue:43441", "accept": false, - "reason": "lm_head weight tying in Mistral3 and FlashAttention failure in Ministral-3 are distinct model-specific bugs." + "reason": "Different Mistral-family bugs: lm_head weight tying vs FlashAttention regression." }, { "left": "issue:42175", "right": "issue:43316", "accept": false, - "reason": "Packaging for the torch extra missing TensorFlow is unrelated to a Gemma3TextConfig API discrepancy." + "reason": "Package dependency/extra install issue vs model config API discrepancy." }, { - "left": "issue:44162", - "right": "issue:44792", + "left": "issue:25251", + "right": "issue:28282", "accept": false, - "reason": "ESM2 breakage and a janus image-generation test failure are different model/test issues." + "reason": "Pipeline output-format bug vs missing PyTorch dependency import error." }, { - "left": "issue:25251", - "right": "issue:28282", + "left": "issue:44162", + "right": "issue:44792", "accept": false, - "reason": "top_k changing pipeline output shape is unrelated to AutoModel requiring PyTorch." + "reason": "ESM2 runtime breakage and janus test failure are unrelated." }, { "left": "issue:43295", "right": "issue:43825", "accept": false, - "reason": "One is a runtime regression in processor.tokenizer/image handling; the other is an incorrect pipeline() error message about translation support." + "reason": "Custom processor/tokenizer regression vs pipeline() error-message/documentation issue." }, { "left": "issue:43673", "right": "issue:44743", "accept": false, - "reason": "Both mention cache/state, but one is missing GenerationMixin cache in chunked_prefill and the other is Qwen3.5 recurrent state reset; different mechanisms." + "reason": "Cache/chunked_prefill issue vs recurrent-state reset in qwen3_5." }, { "left": "issue:16998", "right": "issue:22355", "accept": false, - "reason": "A model_max_length question for DeBERTa-V3 is unrelated to a missing transformers.onnx module import." + "reason": "Model_max_length question vs missing transformers.onnx module import error." } ] }, @@ -9379,8 +9403,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -9434,9 +9458,9 @@ "issue:46050" ], "soft_pairs": [ - "issue:41950|issue:43475", - "issue:44077|issue:44534", "issue:43550|issue:43701", + "issue:44077|issue:44534", + "issue:41950|issue:43475", "issue:43606|issue:43701", "issue:44871|issue:45584", "issue:43388|issue:43688", @@ -9464,208 +9488,214 @@ "issue:44521|issue:45381", "issue:43065|issue:44610" ], - "prepared_review_unit_hash": "6bec5eb546916d879a593c0cf87e159351c3c516402ee56f2714b9f7c8668d85", + "prepared_review_unit_hash": "ae5f45b2d8e49743eeb9560829cf8ee5264eacce28adc252a5271721d15a8bc8", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36415, - "estimated_input_tokens": 9104, - "estimated_eval_tokens": 18464 + "serialized_chars": 36412, + "estimated_input_tokens": 9103, + "estimated_eval_tokens": 18462 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items are broadly related to Transformers bugs, but the soft pairs mostly connect only at a subsystem level rather than the same underlying defect, so I rejected all of them.", - "confidence": 0.89, - "canonical_issue_reason": "Issue 45698 is the broadest active representative of the loading/save-pretrained regression family in this set, with strong discussion activity and a generic failure mode; there is no single true duplicate core across the whole cluster.", + "summary": "The soft pairs are mostly superficial similarity matches across unrelated Transformers bugs; none are true duplicates, and there are no PRs in this cluster.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44521 is the most active open issue with the strongest downstream signal (multiple inbound references), so it is the best representative if one issue must anchor this heterogeneous cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45698 is the best overall anchor because it is active, broadly framed, and closer to several other loading/serialization-related reports than the narrower model-specific bugs.", + "best_issue_reason": "issue:44521 is the most suitable issue to surface for triage because it is open, well-discussed, and externally referenced, even though it is not a duplicate of the other items.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:41950", - "right": "issue:43475", + "left": "issue:43550", + "right": "issue:43701", "accept": false, - "reason": "Both are multimodal/video-related, but one is a processor-selection bug and the other is a missing SAM3 vision output attribute; different code paths." + "reason": "Different bugs: torch.compile/SDPA failure vs resume_from_checkpoint key mismatch." }, { "left": "issue:44077", "right": "issue:44534", "accept": false, - "reason": "One is a model-specific optional post_init policy issue; the other is a framework-wide buffer initialization regression." + "reason": "API/post-init policy issue vs non-persistent buffer initialization regression." }, { - "left": "issue:43550", - "right": "issue:43701", + "left": "issue:41950", + "right": "issue:43475", "accept": false, - "reason": "Unrelated failures: torch.compile/SDPA behavior vs checkpoint key mismatch." + "reason": "Pipeline processor selection bug vs Sam3Video output attribute error; unrelated code paths." }, { "left": "issue:43606", "right": "issue:43701", "accept": false, - "reason": "CPU offload device mismatch is unrelated to a resume_from_checkpoint key mismatch." + "reason": "CPU offload device mismatch vs checkpoint key mismatch; not the same failure." }, { "left": "issue:44871", "right": "issue:45584", "accept": false, - "reason": "Both involve token handling, but the bugs and model families are different." + "reason": "EOS token config inconsistency vs Whisper empty-transcription crash after token alignment." }, { "left": "issue:43388", "right": "issue:43688", "accept": false, - "reason": "Different problems: metric gathering drops labels, while the other is auxiliary-loss normalization." + "reason": "Last-batch metric gathering bug vs MoE auxiliary-loss normalization bug." }, { "left": "issue:43761", "right": "issue:44479", "accept": false, - "reason": "Different models and different regressions; no shared concrete code path." + "reason": "CLIPVision hidden_states regression vs Qwen video-input regression; different models and symptoms." }, { "left": "issue:42617", "right": "issue:43575", "accept": false, - "reason": "A generic execution failure is not the same as an OOM during tensor-parallel loading of a large model." + "reason": "Script/runtime failure vs OOM on large-model tensor-parallel loading." }, { "left": "issue:43295", "right": "issue:44871", "accept": false, - "reason": "Both touch model/processor config, but one is a processor.tokenizer regression and the other is eos_token_id inconsistency." + "reason": "Gemma4 processor/tokenizer regression vs Gemma eos_token_id config mismatch." }, { "left": "issue:43575", "right": "issue:43742", "accept": false, - "reason": "Different load-time failures on different models: OOM vs missing key." + "reason": "OOM on Qwen2-57B load vs MobileLLM config/key error; unrelated load failures." }, { "left": "issue:45127", "right": "issue:45200", "accept": false, - "reason": "One is about weight tying after vocab extension; the other is a missing multimodal token-type default." + "reason": "LoRA merge with tied embeddings vs multimodal token_type_ids defaulting; different mechanisms." }, { "left": "issue:43452", "right": "issue:44496", "accept": false, - "reason": "Both are loading-related, but gguf_file handling is not the same bug as a missing model_type config error." + "reason": "GGUF tokenizer/model loading path bug vs missing model_type in config.json." }, { "left": "issue:44871", "right": "issue:46050", "accept": false, - "reason": "Token-id mismatch and quantized save/load breakage are different issues." + "reason": "EOS config mismatch vs quantized save/load breakage; not the same underlying bug." }, { "left": "issue:43646", "right": "issue:44913", "accept": false, - "reason": "A generic v5 custom-model init break is not the same as rotary_pct not persisting on reload." + "reason": "Custom model initialization regression vs GPTNeoX config reload reverting rotary_pct." }, { "left": "issue:43504", "right": "issue:45698", "accept": false, - "reason": "Both are load/save related, but one is a legacy preset field and the other loads the wrong custom module after save_pretrained." + "reason": "Legacy-field pretrained load issue vs wrong custom module loaded after save_pretrained." }, { "left": "issue:44734", "right": "issue:45406", "accept": false, - "reason": "Distinct serve-time failures in different components: KV-cache indexing vs missing processor tokenizer state." + "reason": "KV-cache continuation indexing crash vs Gemma4Processor missing _tokenizer." }, { "left": "issue:44617", "right": "issue:45584", "accept": false, - "reason": "OOM in Sam3Video is unrelated to Whisper empty-transcription handling." + "reason": "Sam3Video CUDA OOM vs Whisper empty-transcription generation failure." }, { "left": "issue:32090", "right": "issue:35141", "accept": false, - "reason": "Distributed broadcast of None is unrelated to embedding reinitialization after resize." + "reason": "Trainer GPU broadcast of NoneType vs output embedding reinit in post_init." }, { "left": "issue:45563", "right": "issue:45584", "accept": false, - "reason": "A stale warning in paged generate is unrelated to Whisper generation on empty transcription." + "reason": "Stale warning for num_return_sequences vs actual generation failure on empty transcription." }, { "left": "issue:43761", "right": "issue:44242", "accept": false, - "reason": "Hidden_states being None and auxiliary loss omission are different model behaviors." + "reason": "Vision-model hidden_states regression vs load-balancing loss omission in MoE routing." }, { "left": "issue:43525", "right": "issue:43531", "accept": false, - "reason": "Missing pad_token_id and sliding_window behavior are separate config/model bugs." + "reason": "Missing pad_token_id on Llama4Config vs Qwen3-MoE sliding_window handling." }, { "left": "issue:43931", "right": "issue:44060", "accept": false, - "reason": "A Qwen3-VL weight-shape mismatch is not the same as a tied-weights warning bug." + "reason": "Qwen3-VL weight-shape mismatch vs incorrect tied-weights warning in Qwen3-Next." }, { "left": "issue:44792", "right": "issue:44977", "accept": false, - "reason": "Different models and symptoms; one is a test failure, the other is a flash-attention generation regression." + "reason": "Janus image-generation test failure vs Qwen3.5 flash-attention generation bug." }, { "left": "issue:44479", "right": "issue:44734", "accept": false, - "reason": "Video-input regression and KV-cache continuation are unrelated code paths." + "reason": "Video-input regression in Qwen VL models vs serve KV-cache indexing crash." }, { "left": "issue:43906", "right": "issue:45290", "accept": false, - "reason": "No evidence they share the same bug; one is a reproduction of another issue, the other is a tool-call chat-template crash." + "reason": "Repro of an older issue vs assistant-message/tool-call crash; similar area but not clearly the same bug." }, { "left": "issue:44464", "right": "issue:45563", "accept": false, - "reason": "Chunked generation correctness under compilation is unrelated to a stale warning about num_return_sequences." + "reason": "Compiled-forward chunked generation inconsistency vs a stale warning about num_return_sequences." }, { "left": "issue:44190", "right": "issue:44964", "accept": false, - "reason": "Local dataset loading and multimodal model loading are different problems." + "reason": "Local dataset loading script issue vs model-loading failure for Phi-4 multimodal." }, { "left": "issue:44521", "right": "issue:45381", "accept": false, - "reason": "Both are multimodal formatting bugs, but one is assistant_mask handling and the other is wrong vision_position_ids." + "reason": "All-zero assistant masks in multimodal templates vs wrong vision_position_ids for Qwen video input." }, { - "left": "issue:45127", - "right": "issue:45200", + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Multimodal assistant-mask bug vs assistant-message/tool-call crash; related surface area but different concrete failures." + }, + { + "left": "issue:44521", + "right": "issue:45584", "accept": false, - "reason": "Different bugs in post-training/model config behavior; no same underlying change." + "reason": "Multimodal chat-template masking vs Whisper empty-transcription handling; unrelated models and paths." } ] }, @@ -9684,8 +9714,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -9747,8 +9777,8 @@ "issue:44933|issue:45362", "issue:35141|issue:42490", "issue:45127|issue:45357", - "issue:44737|issue:45803", "issue:43335|issue:44855", + "issue:44737|issue:45803", "issue:44462|issue:45907", "issue:43550|issue:44190", "issue:43976|issue:44855", @@ -9770,20 +9800,20 @@ "issue:41950|issue:44016", "issue:36246|issue:38617" ], - "prepared_review_unit_hash": "ca6be969b549ad61ae5ad02a3b0cb4436a1932a64f223d2a643c1353576b49a7", + "prepared_review_unit_hash": "132b057c25b57e9143c110f66208f234b5d6b1425818a62302343e9c1606f74a", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 30, - "serialized_chars": 36921, + "serialized_chars": 36924, "estimated_input_tokens": 9231, "estimated_eval_tokens": 18718 }, @@ -9792,192 +9822,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is noisy and heterogeneous: the soft-similar issue pairs share broad keywords but not the same concrete bug or code path, so none should be merged as duplicates.", - "confidence": 0.95, - "canonical_issue_reason": "No single canonical issue stands out; these items describe unrelated bugs, regressions, docs problems, and import/config errors rather than one underlying defect.", + "summary": "This cluster is mostly a set of unrelated closed/open issues that only share broad topical keywords; I did not find any soft pair that clearly describes the same underlying bug. The closest overlap is in token-embedding resize behavior, but even that does not rise to a safe duplicate merge.", + "confidence": 0.89, + "canonical_issue_reason": "issue:35141 is the cleanest and most general report in the only somewhat related embedding-resize thread: it pinpoints the post_init reinitialization problem when tie_word_embedding=False. The nearby embedding report (issue:45276) is more model-specific and not a confirmed duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "No clear global best issue for this cluster, since the items do not converge on one shared problem.", + "best_issue_reason": "issue:35141 is the best representative issue because it states a concrete, root-cause-level bug in the core resize_token_embeddings path and is more broadly actionable than the more specialized reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43827", "right": "issue:43994", "accept": false, - "reason": "Docs still referencing pipeline() is unrelated to SigLIP2 producing bad outputs with AutoModel/pipeline." + "reason": "Docs still mentioning pipeline() and a SigLIP2 inference mismatch/quantization problem are different bug classes." }, { "left": "issue:44534", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk in v5 is a different bug from GPTNeoX rotary_pct resetting on reload." + "reason": "One is about junk-filled non-persistent buffers in v5; the other is about rotary_pct not persisting on reload." }, { "left": "issue:41720", "right": "issue:45245", "accept": false, - "reason": "Qwen3 device-mapping CUDA assert and category-cardinality limits are unrelated failures." + "reason": "Qwen3 auto device-map CUDA asserts and a categorical limit RuntimeError are unrelated." }, { "left": "issue:44792", "right": "issue:45600", "accept": false, - "reason": "Janus test failure is not the same bug as stale auto_mappings references to removed configs." + "reason": "A Janus image-generation test failure and a stale auto_mappings CI break are different code paths." }, { "left": "issue:44855", "right": "issue:45588", "accept": false, - "reason": "Python 3.13 indentation parsing and flash_attention s_aux=None crashes are different code paths." + "reason": "Python 3.13 import parsing failure in DebertaV2 and a flash-attention sink_aux crash are unrelated." }, { "left": "issue:44933", "right": "issue:45362", "accept": false, - "reason": "Missing image_utils import is unrelated to Qwen3.5 chat crashing." + "reason": "Missing image_utils import and a Qwen3.5 chat crash do not share the same underlying bug." }, { "left": "issue:35141", "right": "issue:42490", "accept": false, - "reason": "Embedding reinitialization on resize is not the same as trailing-slash local-model loading warnings." + "reason": "Token-embedding resize/reinit behavior is unrelated to a local-path trailing-slash warning." }, { "left": "issue:45127", "right": "issue:45357", "accept": false, - "reason": "LoRA+extended-vocab collapse on tied embeddings is different from wrong visual encoder keys in save_pretrained." + "reason": "LoRA merge collapse with extended vocab and incorrect visual encoder keys in save_pretrained are different failures." }, { - "left": "issue:44737", - "right": "issue:45803", + "left": "issue:43335", + "right": "issue:44855", "accept": false, - "reason": "XLNet CPU device placement bug and a bare except swallowing errors are unrelated." + "reason": "SwitchTransformers sparse-layer construction and a Python 3.13 import parsing error are unrelated." }, { - "left": "issue:43335", - "right": "issue:44855", + "left": "issue:44737", + "right": "issue:45803", "accept": false, - "reason": "Sparse-layer creation in SwitchTransformers is unrelated to the Python 3.13 import/parsing failure." + "reason": "XLNet device placement in relative positional encoding and a bare-except bug in Fuyu are different issues." }, { "left": "issue:44462", "right": "issue:45907", "accept": false, - "reason": "Ignoring tokenizer.json and not forwarding revision in list_repo_files are different repository-access bugs." + "reason": "Ignoring tokenizer.json and forwarding list_repo_files revision are separate repository/API bugs." }, { "left": "issue:43550", "right": "issue:44190", "accept": false, - "reason": "torch.compile SDPA failure in Bamba is unrelated to loading a local dataset in the training script." + "reason": "A Bamba torch.compile SDPA failure and a local dataset loading failure in a training script are unrelated." }, { "left": "issue:43976", "right": "issue:44855", "accept": false, - "reason": "Python 3.9+ compatibility failure and Python 3.13 IndentationError are distinct issues." + "reason": "Python version compatibility breakage and a DebertaV2 import indentation/parsing issue are not the same bug." }, { "left": "issue:44077", "right": "issue:44079", "accept": false, - "reason": "Optional post_init handling in patchtsmixer is not the same as ModelOutput key assignment with prior None values." + "reason": "Optional post_init enforcement in patchtsmixer and ModelOutput key assignment with None are different internal behaviors." }, { "left": "issue:44938", "right": "issue:45542", "accept": false, - "reason": "Python 3.14 import/load failure is unrelated to the TensorFlow backend warning from tensorboard-only installs." + "reason": "Python 3.14 load failure and missing TensorFlow backend handling are unrelated packaging/runtime problems." }, { "left": "issue:43644", "right": "issue:44913", "accept": false, - "reason": "Non-persistent buffer junk and GPTNeoX rotary_pct reloading are separate configuration/runtime bugs." + "reason": "Non-persistent buffers filled with junk and rotary_pct resetting on reload are different serialization/config issues." }, { "left": "issue:43540", "right": "issue:44734", "accept": false, - "reason": "Qwen3OmniMoe video-input validation and serve KV-cache tensor indexing are different failures." + "reason": "Qwen3OmniMoe video-input validation and serve KV-cache tensor indexing are distinct code paths." }, { "left": "issue:44112", "right": "issue:45600", "accept": false, - "reason": "A stale GraniteSpeech CI test is not the same as removed Sam3LiteText config references breaking CI." - }, - { - "left": "issue:43335", - "right": "issue:44190", - "accept": false, - "reason": "SwitchTransformers sparse-layer initialization and local dataset loading are unrelated." + "reason": "GraniteSpeech CI flakiness from stale device overrides is unrelated to removed Sam3LiteText config references." }, { "left": "issue:46082", "right": "issue:46121", "accept": false, - "reason": "Llama head_dim validation and rope-validation list handling are different parameter-validation bugs." + "reason": "LlamaConfig head_dim validation and rope-parameter JSON list handling are separate configuration bugs." }, { "left": "issue:39692", "right": "issue:44190", "accept": false, - "reason": "SigLIP2 docs example errors are unrelated to the image-classification local dataset loading issue." + "reason": "SigLIP2 documentation errors and a dataset-loading failure in a training script are not duplicates." }, { "left": "issue:44496", "right": "issue:45803", "accept": false, - "reason": "Unrecognized model/config loading and a bare except swallowing errors are not the same bug." + "reason": "An unrecognized-model/config loading error and a bare-except bug are unrelated." }, { "left": "issue:36246", "right": "issue:41628", "accept": false, - "reason": "These are different import/export failures for different symbols and modules, not one shared defect." + "reason": "Both are import errors, but they involve different missing symbols and modules, not the same export bug." }, { "left": "issue:45276", "right": "issue:45584", "accept": false, - "reason": "Gemma4 resize_token_embeddings not updating embeddings is unrelated to Whisper empty-transcription generation." + "reason": "Gemma4 embedding-resize behavior and Whisper empty-transcription generation failure are unrelated." }, { "left": "issue:43479", "right": "issue:45375", "accept": false, - "reason": "Phi4Multimodal None-default initialization and missing Qwen3.5MoeVisionConfig fields are different config bugs." + "reason": "Default multimodal config initialization and a missing deepstack_visual_indexes field are different config problems." }, { "left": "issue:45245", "right": "issue:45584", "accept": false, - "reason": "Category-count limit runtime error is unrelated to Whisper generation after align_special_tokens." + "reason": "A category-count overflow and a Whisper empty-transcription bug are unrelated." }, { "left": "issue:43650", "right": "issue:44220", "accept": false, - "reason": "A vague placeholder issue title does not match the specific _torch_extract_fbank_features problem." + "reason": "The placeholder title 'ADD THE DATA' provides no overlap with the fbank feature extraction bug." }, { "left": "issue:43295", "right": "issue:45276", "accept": false, - "reason": "Processor/tokenizer regression in v4.57.5 is unrelated to Gemma4 token-embedding resize behavior." + "reason": "Processor/tokenizer regression and Gemma4 resize_token_embeddings behavior concern different multimodal and embedding paths." }, { "left": "issue:41950", "right": "issue:44016", "accept": false, - "reason": "Video-classification processor lookup and a notebook syntax error are different problem types." + "reason": "A video-classification pipeline processor lookup issue and a notebook syntax error are unrelated." }, { "left": "issue:36246", "right": "issue:38617", "accept": false, - "reason": "These are separate import errors from different modules and symbols, not the same underlying bug." + "reason": "These are separate import errors for different symbols in different modules." } ] }, @@ -9996,8 +10020,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -10053,9 +10077,9 @@ "issue:44186|issue:44734", "issue:42175|issue:44279", "issue:44683|issue:45446", - "issue:44016|issue:44220", "issue:38175|issue:43065", "issue:44246|issue:45003", + "issue:44016|issue:44220", "issue:43404|issue:45216", "issue:44734|issue:45632", "issue:44561|issue:45632", @@ -10063,8 +10087,8 @@ "issue:44964|issue:45600", "issue:44112|issue:44964", "issue:43854|issue:44964", - "issue:42907|issue:43441", "issue:43976|issue:44987", + "issue:42907|issue:43441", "issue:42490|issue:45632", "issue:43531|issue:45600", "issue:43673|issue:44336", @@ -10083,20 +10107,20 @@ "issue:44496|issue:45759", "issue:43299|issue:45357" ], - "prepared_review_unit_hash": "ea10d663e6d0236261daa2d215be7bfc612e0f97ed83ee77bbc54e090bed25f4", + "prepared_review_unit_hash": "6d0803fc46f7affd8d77bbfe73d9b0bb9d64a0818ca8440f9f1f6b8f358dc1ad", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 47, "item_count": 47, "soft_pair_count": 32, - "serialized_chars": 36656, + "serialized_chars": 36653, "estimated_input_tokens": 9164, "estimated_eval_tokens": 18584 }, @@ -10105,204 +10129,246 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The cluster is a heterogeneous mix of unrelated Transformers issues; none of the soft-edge pairs look like true duplicates or mergeable PR-equivalents. I\u2019d use the broadly discussed import-time/performance report as the best representative issue.", - "confidence": 0.94, - "canonical_issue_reason": "issue:44246 is the strongest representative overall: it has the most discussion activity, an inbound reference, and a broad user-facing import-performance symptom that is easy to map to the cluster\u2019s general maintenance/debugging theme.", + "summary": "This cluster is heterogeneous: the items cover unrelated regressions across model loading, tokenizers, save/load behavior, attention, docs, and CI. None of the soft-edge pairs look like the same concrete bug/change, so I would not merge any of them as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue stands out because the set spans many different failure modes and subsystems rather than one underlying bug.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44246 stands out as the most canonical issue because it is widely discussed, externally referenced, and describes a concrete, broadly relevant regression rather than a narrow model-specific edge case.", + "best_issue_reason": "No strong global best issue; if forced, a concrete repro-oriented model-loading report would be the most representative, but the cluster is too mixed for a true canonical.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44186", "right": "issue:44734", "accept": false, - "reason": "Different failures: LayoutLMv2 tokenizer batching/NER crash vs server-side KV-cache continuation indexing bug." + "reason": "Different code paths: LayoutLMv2 NER/padding crash versus serve KV-cache continuation indexing." }, { "left": "issue:42175", "right": "issue:44279", "accept": false, - "reason": "One is a backend/install packaging omission; the other is an unspecific dependency complaint with no clear shared code path." + "reason": "One is a missing TensorFlow backend in installation extras; the other is a vague dependency problem." }, { "left": "issue:44683", "right": "issue:45446", "accept": false, - "reason": "Both touch flex_attention, but they describe different version-gating problems and different failure points." - }, - { - "left": "issue:44016", - "right": "issue:44220", - "accept": false, - "reason": "Notebook syntax error vs audio feature extraction issue; unrelated artifacts." + "reason": "Both mention flex_attention, but one is a torch>=2.9 compilation failure and the other is a version-check/import bug for AuxRequest." }, { "left": "issue:38175", "right": "issue:43065", "accept": false, - "reason": "SigLIP zero-probabilities bug and Sam3PixelDecoder dummy Conv2d are unrelated model implementations." + "reason": "Unrelated models and failures: SigLIP2 zero probabilities versus a dummy Conv2d in Sam3PixelDecoder." }, { "left": "issue:44246", "right": "issue:45003", "accept": false, - "reason": "Import slowdown and unsafe sys.modules access may be related in theme, but they do not clearly describe the same bug." + "reason": "Import slowness and unsafe sys.modules access may be adjacent, but they are not the same concrete bug." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "A notebook syntax error is unrelated to the _torch_extract_fbank_features issue." }, { "left": "issue:43404", "right": "issue:45216", "accept": false, - "reason": "Different models and different symptoms: weight tying bug vs incorrect save_pretrained checkpoint serialization." + "reason": "Different bugs: untied lm_head weights in Mistral3 versus incorrect save_pretrained output for Qwen3.5." }, { "left": "issue:44734", "right": "issue:45632", "accept": false, - "reason": "KV-cache continuation crash is unrelated to trust_remote_code cache-path collision." + "reason": "Serve KV-cache continuation crash is unrelated to local-model cache path collisions with trust_remote_code." }, { "left": "issue:44561", "right": "issue:45632", "accept": false, - "reason": "Removal of is_torch_fx_available breaking remote-code models is a different compatibility break than local-model cache collisions." + "reason": "Removal of is_torch_fx_available affecting trust_remote_code is a different regression from local-model cache path collisions." }, { "left": "issue:43643", "right": "issue:44496", "accept": false, - "reason": "Missing fields from trust_remote_code AutoConfig return object is a different config-loading bug than unrecognized model_type handling." + "reason": "Missing fields from trust_remote_code AutoConfig and an unrecognized-model/model_type error are different loading failures." }, { "left": "issue:44964", "right": "issue:45600", "accept": false, - "reason": "Phi-4 multimodal loading failure and stale auto_mappings references are distinct breakages." + "reason": "Phi-4 multimodal loading failure is unrelated to stale auto_mappings references to Sam3LiteText configs." }, { "left": "issue:44112", "right": "issue:44964", "accept": false, - "reason": "CI test failure in GraniteSpeech is not the same underlying issue as Phi-4 multimodal loading." + "reason": "CI stale device override test failure is not the same as model loading for Phi-4 multimodal." }, { "left": "issue:43854", "right": "issue:44964", "accept": false, - "reason": "GLM-4.7-Flash load failure and Phi-4 multimodal load failure are separate model-specific issues." + "reason": "Different model families and failure modes: GLM-4.7-Flash unit-test loading versus Phi-4 multimodal loading." }, { - "left": "issue:42907", - "right": "issue:43441", + "left": "issue:43976", + "right": "issue:44987", "accept": false, - "reason": "Saving dequantized Ministral/Devstral models is unrelated to FlashAttention loading failures." + "reason": "Python version compatibility problem is unrelated to loading physical-intelligence/fast." }, { - "left": "issue:43976", - "right": "issue:44987", + "left": "issue:42907", + "right": "issue:43441", "accept": false, - "reason": "Python version support regression and model-loading failure are unrelated." + "reason": "Saving dequantized Ministral/Devstral models is different from a FlashAttention failure in Ministral-3." }, { "left": "issue:42490", "right": "issue:45632", "accept": false, - "reason": "Trailing-slash local model warning is a different path-handling bug than leaf-directory cache collisions." + "reason": "A warning for local paths ending with a slash is not the same bug as cache-path collisions for leaf-directory names." }, { "left": "issue:43531", "right": "issue:45600", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and removed Sam3LiteText config references are unrelated." + "reason": "Qwen3-MoE sliding_window behavior and Sam3LiteText auto_mappings cleanup are unrelated." }, { "left": "issue:43673", "right": "issue:44336", "accept": false, - "reason": "Generation cache/chunked_prefill failure has nothing to do with ANSI codes in loading_report." + "reason": "Generation cache missing during chunked prefill is unrelated to ANSI codes in loading_report." }, { "left": "issue:43295", "right": "issue:45335", "accept": false, - "reason": "Processor/tokenizer regression is unrelated to t5gemma decoder embedding resize behavior." + "reason": "Processor/tokenizer API regression is unrelated to t5gemma resize_token_embeddings not updating decoder.embed_tokens." }, { "left": "issue:44016", "right": "issue:44062", "accept": false, - "reason": "Notebook syntax error and AddedToken argument collision are unrelated." + "reason": "Notebook syntax error and AddedToken keyword conflict are different issues." }, { "left": "issue:43526", "right": "issue:43901", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug and pipeline docs mismatch are different layers and symptoms." + "reason": "BeitImageProcessorFast label reduction bug is not the same as stale pipeline docs about return_all_scores." }, { "left": "issue:43824", "right": "issue:44933", "accept": false, - "reason": "Both are import-related, but they concern different missing exports and different modules." + "reason": "Missing Qwen2_5_VLForConditionalGeneration import is different from a non-existent image_utils import." }, { "left": "issue:44496", "right": "issue:45042", "accept": false, - "reason": "Model-type recognition failure and unintended torchvision dependency are separate issues." + "reason": "Unrecognized model/config loading and PIL image processors requiring torchvision are unrelated failures." }, { "left": "issue:43335", "right": "issue:45593", "accept": false, - "reason": "SwitchTransformers sparse-layer config bug and D-FINE auxiliary-loss behavior are unrelated." + "reason": "Sparse layer creation in SwitchTransformersConfig is unrelated to D-FINE auxiliary-loss behavior." }, { "left": "issue:43299", "right": "issue:45216", "accept": false, - "reason": "Qwen3VL model-loading regression and Qwen3.5 save_pretrained regression are different code paths." + "reason": "Qwen3VL MoE model loading and Qwen3.5 save_pretrained checkpoint corruption are different bugs." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Optimizer SGD arguments not being used is unrelated to tie_word_embeddings warnings in Qwen3.5 LoRA fine-tuning." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python 3.9+/3.10+ compatibility is not the same as a general dependency issue." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are different code paths." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Documentation drift for text classification is unrelated to unrecognized-model config loading." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "T5Gemma embedding resize bug is unrelated to Whisper empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized-model/config.json issues are different from AutoModelForCausalLM.from_config failing to unwrap text_config." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Loading failure for Qwen3VL MoE and incorrect saved visual encoder keys for Qwen3.5 are different regressions." }, { "left": "issue:44075", "right": "issue:44368", "accept": false, - "reason": "SGD optimizer args being ignored is unrelated to tie_word_embeddings warnings." + "reason": "SGD argument handling is unrelated to warning about tie_word_embeddings in Qwen3.5 LoRA." }, { "left": "issue:43976", "right": "issue:44279", "accept": false, - "reason": "Python compatibility issue is not the same as a generic dependency issue." + "reason": "Python version support and dependency installation problems are not the same bug." }, { "left": "issue:43381", "right": "issue:45593", "accept": false, - "reason": "Gradient-checkpointing eval-mode restriction is unrelated to D-FINE auxiliary-loss omission." + "reason": "Eval-mode gradient checkpointing failure and D-FINE auxiliary-loss omission do not share a concrete fix." }, { "left": "issue:43901", "right": "issue:44496", "accept": false, - "reason": "Docs/behavior mismatch for return_all_scores is unrelated to unrecognized model configs." + "reason": "A docs mismatch in TextClassificationPipeline is unrelated to config/model-type loading errors." }, { "left": "issue:45335", "right": "issue:45584", "accept": false, - "reason": "Embedding resize bug and Whisper empty-transcription failure are different problems." + "reason": "Embedding resize behavior in t5gemma does not overlap with Whisper generation on empty transcription." }, { "left": "issue:44496", "right": "issue:45759", "accept": false, - "reason": "Unrecognized model_type handling and text_config unwrapping for composite configs are separate loading paths." + "reason": "These are distinct multimodal/config loading issues with different failure points." }, { "left": "issue:43299", "right": "issue:45357", "accept": false, - "reason": "Both are Qwen3.5-related, but one is a load-time regression and the other is incorrect save_pretrained visual-encoder keys." + "reason": "Model loading failure and save_pretrained key regression are different underlying issues." } ] }, @@ -10321,8 +10387,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -10376,14 +10442,14 @@ "issue:46132" ], "soft_pairs": [ - "issue:41950|issue:45020", "issue:44242|issue:45200", + "issue:41950|issue:45020", "issue:44351|issue:45278", - "issue:44521|issue:45325", "issue:43479|issue:43525", + "issue:44521|issue:45325", "issue:44683|issue:45542", - "issue:41950|issue:45823", "issue:44779|issue:45812", + "issue:41950|issue:45823", "issue:43976|issue:45405", "issue:43577|issue:45216", "issue:38175|issue:44617", @@ -10405,202 +10471,202 @@ "issue:44466|issue:45276", "issue:43065|issue:44898" ], - "prepared_review_unit_hash": "a9ad59a5ffd7aa85815da97d2129dc2f731657e10ea297400041760d8c7c561f", + "prepared_review_unit_hash": "f54fb0940a09f0cfc644997dcea4e680bedd5d0c94019dc2d148f45d183129d7", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 28, - "serialized_chars": 36035, - "estimated_input_tokens": 9009, - "estimated_eval_tokens": 18274 + "serialized_chars": 36032, + "estimated_input_tokens": 9008, + "estimated_eval_tokens": 18272 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is very broad and mostly contains unrelated issue reports across tokenizers, multimodal configs, docs, deepspeed, quantization, and version regressions. None of the soft edges look like the same underlying bug/change, so I would not merge any of them as duplicates.", - "confidence": 0.83, - "canonical_issue_reason": "issue:45020 is the broadest umbrella-style report in the set ('Recent transformers versions break models using remote_code') and is the most plausible anchor for a regression cluster.", + "summary": "No soft pair looks like a true duplicate; the cluster is mostly a set of unrelated model, tokenizer, config, and infrastructure bugs that only share broad Transformers/Deepspeed similarity.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44521 is the best fallback representative because it is a concrete, actionable bug with strong discussion/inbound linkage, even though it is not a duplicate of the other items.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45020 is the best representative issue for the shared high-level theme of recent-version regressions, even though the rest of the items are not true duplicates of it.", + "best_issue_reason": "issue:44521 is the clearest overall anchor in this set: active, specific, and more representative than the broader regression reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:41950", - "right": "issue:45020", + "left": "issue:44242", + "right": "issue:45200", "accept": false, - "reason": "Different problems: video-classification processor lookup vs remote_code model-loading regressions." + "reason": "Different bugs: MoE load-balancing loss vs Gemma4 multimodal config defaults." }, { - "left": "issue:44242", - "right": "issue:45200", + "left": "issue:41950", + "right": "issue:45020", "accept": false, - "reason": "Different model behavior bugs: MoE router loss handling vs Gemma4 text-only token type defaults." + "reason": "Video-classification processor lookup and remote_code breakage are unrelated failures." }, { "left": "issue:44351", "right": "issue:45278", "accept": false, - "reason": "One is a specific HybridCache import failure; the other is a broad import-error report after upgrade." + "reason": "One is a missing HybridCache import; the other is a broad import-regression report." }, { - "left": "issue:44521", - "right": "issue:45325", + "left": "issue:43479", + "right": "issue:43525", "accept": false, - "reason": "Different multimodal bugs: assistant mask generation vs Qwen2.5-VL rope position scaling." + "reason": "Different config issues: Phi4 multimodal defaults vs missing pad_token_id on Llama4Config." }, { - "left": "issue:43479", - "right": "issue:43525", + "left": "issue:44521", + "right": "issue:45325", "accept": false, - "reason": "Different config issues: missing default vision/audio config initialization vs absent pad_token_id on Llama4Config." + "reason": "Multimodal assistant masks and Qwen2.5-VL rope indexing are different code paths." }, { "left": "issue:44683", "right": "issue:45542", "accept": false, - "reason": "Unrelated: compiled flex_attention on torch 2.9 vs TensorBoard/TensorFlow backend installation problem." + "reason": "FlexAttention compilation on torch 2.9 is unrelated to a TensorFlow/tensorboard backend error." }, { - "left": "issue:41950", - "right": "issue:45823", + "left": "issue:44779", + "right": "issue:45812", "accept": false, - "reason": "Both are Gemma/vision-related only at a very broad level; the concrete failures are different." + "reason": "Both mention tokenizer regressions, but they affect different model families and likely different paths." }, { - "left": "issue:44779", - "right": "issue:45812", + "left": "issue:41950", + "right": "issue:45823", "accept": false, - "reason": "Both involve tokenizers, but they affect different models and different failure modes." + "reason": "Video pipeline processor lookup is not the same as Gemma4 device-mismatch during forward." }, { "left": "issue:43976", "right": "issue:45405", "accept": false, - "reason": "Python version compatibility issue vs dependency version pinning mismatch; not the same bug." + "reason": "Python-version compatibility and PEFT version pinning are separate dependency problems." }, { "left": "issue:43577", "right": "issue:45216", "accept": false, - "reason": "Different regressions: dtype propagation in Blip2 vs incorrect checkpoint serialization in Qwen3.5." + "reason": "Blip2 dtype loading and Qwen3.5 save_pretrained serialization are different bugs." }, { "left": "issue:38175", "right": "issue:44617", "accept": false, - "reason": "Completely different symptoms: zero probabilities in SigLIP2 vs CUDA OOM in Sam3Video." + "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to a Sam3Video OOM." }, { "left": "issue:44279", "right": "issue:45405", "accept": false, - "reason": "Generic dependency problem vs a specific unreleased PEFT minimum-version bump." + "reason": "A generic dependency issue is not the same as an unreleased PEFT minimum-version bump." }, { "left": "issue:39692", "right": "issue:43901", "accept": false, - "reason": "Both touch docs/examples, but one is a SigLIP2 example bug and the other is outdated pipeline docs." + "reason": "Documentation issues, but for different examples, APIs, and failure modes." }, { "left": "issue:43299", "right": "issue:43854", "accept": false, - "reason": "Different model-loading failures for different models; no shared code-path is evident." + "reason": "Different model-loading failures for different multimodal models." }, { "left": "issue:43650", "right": "issue:46132", "accept": false, - "reason": "The titles do not describe the same bug or change; one is vague, the other is an AttentionInterface behavior change." + "reason": "A placeholder title is unrelated to an AttentionInterface.register behavior change." }, { "left": "issue:43295", "right": "issue:44779", "accept": false, - "reason": "Different regressions: processor.tokenizer custom-code breakage vs Deepseek tokenizer output changes." + "reason": "Processor/tokenizer regression and Deepseek tokenizer output regression are not the same bug." }, { "left": "issue:44462", "right": "issue:45081", "accept": false, - "reason": "AutoTokenizer ignoring repo tokenizer.json is not the same as Mistral regex patch crashing on Tokenizer internals." + "reason": "Ignoring tokenizer.json and crashing in _patch_mistral_regex are separate tokenizer-loading issues." }, { "left": "issue:45081", "right": "issue:45907", "accept": false, - "reason": "Tokenizer patch crash vs repo file listing not forwarding revision; unrelated code paths." + "reason": "A Mistral regex crash is unrelated to list_repo_files failing to forward revision." }, { "left": "issue:44297", "right": "issue:44462", "accept": false, - "reason": "Tokenizer save_pretrained metadata mismatch vs AutoTokenizer loading the wrong tokenizer source." + "reason": "Tokenizer class mismatch on save and tokenizer.json being ignored are different save/load defects." }, { "left": "issue:44016", "right": "issue:44291", "accept": false, - "reason": "Notebook syntax error vs init_empty_weights constructor argument regression; unrelated." + "reason": "Notebook syntax error and init_empty_weights TypeError are unrelated." }, { "left": "issue:41950", "right": "issue:43650", "accept": false, - "reason": "No concrete overlap: video-classification processor bug vs unrelated placeholder issue." + "reason": "Unrelated video-pipeline bug and placeholder issue title." }, { "left": "issue:43450", "right": "issue:44279", "accept": false, - "reason": "Video processor batching shape bug vs generic dependency issue; not the same underlying defect." + "reason": "Video batch-shape bug is unrelated to a dependency issue." }, { "left": "issue:43441", "right": "issue:45834", "accept": false, - "reason": "Different model-specific inference bugs: FlashAttention failure vs OCR long-input index error." + "reason": "FlashAttention failure in Ministral and long-ocr index errors in Kosmos2.5 are different model-specific bugs." }, { "left": "issue:45600", "right": "issue:45712", "accept": false, - "reason": "Both are repo-maintenance/CI issues, but they involve different files and different broken references." + "reason": "Both touch stale code paths, but they concern different files and different CI failures." }, { "left": "issue:33290", "right": "issue:45656", "accept": false, - "reason": "Deepspeed OOM with Adafactor is a different symptom from optimizer.step being called twice." + "reason": "Adafactor OOM under DeepSpeed and double optimizer steps under DeepSpeed are not the same defect." }, { "left": "issue:42994", "right": "issue:43317", "accept": false, - "reason": "Saving a quantized model and loading a dequantized model with device_map=auto are different workflows and failures." + "reason": "Quantized saving failure and dequantized model loading/offload failure are distinct quantization bugs." }, { "left": "issue:44466", "right": "issue:45276", "accept": false, - "reason": "Serialization of lm_head.weight is distinct from resize_token_embeddings not updating Gemma4 embeddings/output heads." + "reason": "lm_head serialization inconsistency and Gemma4 embedding resizing are different weight-handling issues." }, { "left": "issue:43065", "right": "issue:44898", "accept": false, - "reason": "Sam3PixelDecoder dummy Conv2d issue does not match Perceiver non-default resolution failure." + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to Perceiver resolution/interpolation failure." } ] }, @@ -10619,8 +10685,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -10676,13 +10742,13 @@ "soft_pairs": [ "issue:43650|issue:45290", "issue:43653|issue:45593", - "issue:43441|issue:44683", "issue:44623|issue:44749", + "issue:43441|issue:44683", "issue:44261|issue:44263", "issue:44623|issue:45397", "issue:43723|issue:44162", - "issue:42175|issue:43976", "issue:44016|issue:45823", + "issue:42175|issue:43976", "issue:44361|issue:45593", "issue:43262|issue:44521", "issue:43673|issue:43761", @@ -10704,20 +10770,20 @@ "issue:42491|issue:43299", "issue:44242|issue:45593" ], - "prepared_review_unit_hash": "37a4260bb630e6fba00f9c8aa45ada2815caa7bd2944bbb1c9dbcce295b9a91e", + "prepared_review_unit_hash": "7b4a5cc65af13c23322aa61f157d0270066e83cd7b0ac989e6098f9802ddb17d", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 36071, + "serialized_chars": 36072, "estimated_input_tokens": 9018, "estimated_eval_tokens": 18292 }, @@ -10726,186 +10792,186 @@ "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These are mostly unrelated single-bug issues, not true duplicates. None of the soft pairs look like the same underlying bug or change, so all are rejected. The best standalone issue to use as a canonical tracking target is the open multimodal `apply_chat_template` regression.", - "confidence": 0.84, - "canonical_issue_reason": "issue:44521 is the clearest, most actionable bug report in the set: it is specific, reproducible, actively discussed, and describes a concrete regression in `apply_chat_template` for multimodal inputs.", + "summary": "The set is mostly a grab bag of unrelated Transformers bugs; none of the soft pairs are close enough to treat as true duplicates, so I rejected them all. I picked the most substantive/high-signal issue as the canonical representative.", + "confidence": 0.34, + "canonical_issue_reason": "issue:44521 has the clearest, most general bug report with ongoing discussion and multiple references, and it is the strongest representative issue in this mixed set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44521 has the strongest combination of clear scope, user impact, and reproducibility, making it the best canonical issue candidate among this set.", + "best_issue_reason": "issue:44521 is the best issue candidate because it is a concrete, user-impacting regression in a core API (`apply_chat_template`) and has the most evidence of active relevance.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43650", "right": "issue:45290", "accept": false, - "reason": "Unrelated topics: a vague data request vs a crash in chat template handling with tool calls." + "reason": "Unrelated topics: vague data request vs chat template crash on tool-call assistant messages." }, { "left": "issue:43653", "right": "issue:45593", "accept": false, - "reason": "Tokenizer special-token registration bug vs missing auxiliary losses in D-FINE; different code paths." + "reason": "Different models and bugs: tokenizer special-token registration vs missing auxiliary losses in D-FINE." }, { - "left": "issue:43441", - "right": "issue:44683", + "left": "issue:44623", + "right": "issue:44749", "accept": false, - "reason": "Both mention attention kernels, but one is a model-specific FlashAttention failure and the other is a torch-version flex_attention regression." + "reason": "Different problem class: missing saved processor files vs a performance regression after upgrade." }, { - "left": "issue:44623", - "right": "issue:44749", + "left": "issue:43441", + "right": "issue:44683", "accept": false, - "reason": "Saving processor files is unrelated to a performance slowdown in data filtering." + "reason": "Both mention attention kernels, but they target different failure modes and code paths (FlashAttention/model loading vs compiled flex_attention on newer torch)." }, { "left": "issue:44261", "right": "issue:44263", "accept": false, - "reason": "Different bugs in different models: layernorm epsilon handling vs `torch.split()` return values." + "reason": "Different GLM/MoE internals; no clear shared underlying bug." }, { "left": "issue:44623", "right": "issue:45397", "accept": false, - "reason": "Processor save/pretrained file emission is not the same as Gemma4 zero3 loading failure." + "reason": "Processor save file omission vs gemma-4 from_pretrained failure are distinct issues." }, { "left": "issue:43723", "right": "issue:44162", "accept": false, - "reason": "Generic tokenizer loading issue vs ESM2 model breakage; not the same underlying defect." + "reason": "Both are loading-related, but one is generic tokenizer loading and the other is an ESM2 model regression; not the same bug." }, { - "left": "issue:42175", - "right": "issue:43976", + "left": "issue:44016", + "right": "issue:45823", "accept": false, - "reason": "Backend dependency packaging vs Python version support; distinct failures." + "reason": "Notebook syntax error is unrelated to Gemma4 device-mapping failure." }, { - "left": "issue:44016", - "right": "issue:45823", + "left": "issue:42175", + "right": "issue:43976", "accept": false, - "reason": "Notebook syntax error is unrelated to a Gemma4 device-map runtime bug." + "reason": "Packaging/backends problem vs Python-version compatibility regression; different root causes." }, { "left": "issue:44361", "right": "issue:45593", "accept": false, - "reason": "Tokenizer attribute failure vs missing auxiliary losses; different subsystems." + "reason": "Tokenizer task AttributeError and D-FINE auxiliary-loss bug are unrelated." }, { "left": "issue:43262", "right": "issue:44521", "accept": false, - "reason": "Both involve chat templates, but one is audio sampling-rate defaulting and the other is zero assistant masks for multimodal inputs." + "reason": "Same API surface (`apply_chat_template`) but different bugs: audio sampling-rate default vs assistant mask generation." }, { "left": "issue:43673", "right": "issue:43761", "accept": false, - "reason": "Generation cache regression and CLIPVision hidden-states regression are separate model/runtime bugs." + "reason": "Generation cache issue vs CLIPVision hidden_states regression; different models and code paths." }, { "left": "issue:44671", "right": "issue:44779", "accept": false, - "reason": "CamemBERT masked-LM predictions and Deepseek tokenizer output are unrelated model-specific regressions." + "reason": "Different model families and symptoms: CamemBERT MLM predictions vs Deepseek tokenizer output." }, { "left": "issue:44987", "right": "issue:45542", "accept": false, - "reason": "Model loading failure for a specific package is not the same as missing TensorFlow backend installation." + "reason": "Model-loading failure for a specific repo vs missing TensorFlow backend install issue." }, { "left": "issue:43901", "right": "issue:43994", "accept": false, - "reason": "A docs mismatch about `return_all_scores` is unrelated to incorrect SigLIP2 inference results." + "reason": "Docs mismatch in a pipeline API vs incorrect SigLIP2 inference results." }, { "left": "issue:43784", "right": "issue:45542", "accept": false, - "reason": "Importing sentence-transformers fails due to a missing symbol, which is unrelated to backend dependency selection." + "reason": "Import-time NameError in sentence-transformers integration vs TensorFlow backend being absent." }, { "left": "issue:43824", "right": "issue:43976", "accept": false, - "reason": "Missing Qwen2.5-VL import and Python version incompatibility are different issues." + "reason": "Import error for a specific VL class vs Python-version support regression; not the same bug." }, { "left": "issue:43450", "right": "issue:44617", "accept": false, - "reason": "Batched video shape bug is not the same as CUDA OOM in Sam3Video." + "reason": "Video processor batching shape bug vs CUDA OOM for Sam3Video." }, { "left": "issue:43525", "right": "issue:43881", "accept": false, - "reason": "Missing `pad_token_id` on Llama4Config and a GLM-4V loading failure are not the same code-path problem." + "reason": "Different model load failures with no evidence of a shared code-path problem." }, { "left": "issue:44297", "right": "issue:45397", "accept": false, - "reason": "Tokenizer config serialization mismatch is unrelated to Gemma4 zero3 loading." + "reason": "Tokenizer save metadata mismatch vs gemma-4 zero3 loading failure are unrelated." }, { "left": "issue:44188", "right": "issue:44336", "accept": false, - "reason": "Attention-kernel divergence under torch.compile and ANSI-code output in loading_report are unrelated." + "reason": "Attention-kernel divergence vs ANSI escape output in loading_report; unrelated." }, { "left": "issue:43525", "right": "issue:43643", "accept": false, - "reason": "A specific config attribute bug is not the same as missing fields from `trust_remote_code=True` config loading." + "reason": "Both involve missing config fields, but one is a specific model config attr error and the other is a remote-code config loading bug; too different to merge." }, { "left": "issue:41628", "right": "issue:43976", "accept": false, - "reason": "Missing `AutoImageProcessor` import and Python 3.9/3.10 support are unrelated." + "reason": "Import error for a class export vs Python 3.9/3.10 compatibility issue." }, { "left": "issue:44589", "right": "issue:45362", "accept": false, - "reason": "Storage-class lookup failure and Qwen3.5 chat crash are separate runtime issues." + "reason": "Float8 storage lookup error and Qwen3.5 chat crash are unrelated." }, { "left": "issue:43901", "right": "issue:44246", "accept": false, - "reason": "Docs wording and import-time performance are not the same bug." + "reason": "Documentation issue vs intermittent slow import; no shared underlying bug." }, { "left": "issue:45561", "right": "issue:45563", "accept": false, - "reason": "Pytest-xdist file race and a stale generation warning are unrelated." + "reason": "xdist race in testing utils vs stale warning in paged generate; different areas." }, { "left": "issue:44336", "right": "issue:44373", "accept": false, - "reason": "ANSI escape output and a wrong docstring concern different parts of the codebase." + "reason": "Loading-report ANSI formatting bug vs wrong docstring; not the same issue." }, { "left": "issue:42491", "right": "issue:43299", "accept": false, - "reason": "Both mention Qwen3 MoE, but one is a LoRA compatibility regression and the other is a model-loading regression; too different to merge." + "reason": "Both involve Qwen3 MoE loading regressions, but one is a LoRA compatibility problem and the other is a base-model loading breakage; too different for one PR." }, { "left": "issue:44242", "right": "issue:45593", "accept": false, - "reason": "Load-balancing loss gating and D-FINE auxiliary-loss behavior are different model-training issues." + "reason": "Missing load-balancing loss in MoE routing vs D-FINE auxiliary-loss omission; similar wording, different concrete bugs." } ] }, @@ -10924,8 +10990,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -10995,14 +11061,14 @@ "issue:44242|issue:44857", "issue:36246|issue:43596", "issue:44162|issue:45200", - "issue:45412|issue:46123", "issue:44295|issue:44749", + "issue:45412|issue:46123", "issue:44779|issue:45292", "issue:43901|issue:45003", "issue:45292|issue:45584", - "issue:45987|issue:46082", "issue:44496|issue:45003", "issue:45292|issue:45701", + "issue:45987|issue:46082", "issue:44987|issue:45459", "issue:44877|issue:45375", "issue:43232|issue:44336", @@ -11011,220 +11077,220 @@ "issue:43262|issue:44246", "issue:44995|issue:45200" ], - "prepared_review_unit_hash": "283b248fb8a4e6a009bce671037a9bacc0da7611a391c47642be95c89c836d69", + "prepared_review_unit_hash": "45e97d3b85f02b9481c7a266dac8609ad8188766733eb2e3de503a5f7b738ae4", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 31, - "serialized_chars": 36859, - "estimated_input_tokens": 9215, - "estimated_eval_tokens": 18686 + "serialized_chars": 36856, + "estimated_input_tokens": 9214, + "estimated_eval_tokens": 18684 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a set of unrelated Transformers bug reports that were grouped by superficial keyword similarity, not by a shared underlying defect. No soft-edge pair is close enough to merge, and there are no PRs in the input.", - "confidence": 0.92, - "canonical_issue_reason": "No exact duplicate anchor stands out. If one representative issue must be chosen, issue 44246 is the broadest and most discussion-heavy, but it is still not a duplicate of the rest.", + "summary": "This cluster does not look like a true duplicate set; the soft pairs mostly connect unrelated model-loading, config, pipeline, cache, and infrastructure issues. I would not merge any of the proposed soft edges.", + "confidence": 0.96, + "canonical_issue_reason": "issue:44246 is the best cluster anchor because it has the broadest symptom description and high discussion activity, making it the most umbrella-like issue among otherwise heterogeneous reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 44246 is the best cluster representative because it is generic, high-discussion, and closest to the accidental 'import/loading' theme of the group, though it should not be treated as the same bug as the others.", + "best_issue_reason": "issue:44246 is the strongest single issue to represent the set: it is generic, high-activity, and closer to a cross-cutting symptom than the many model-specific failures here.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44295", "right": "issue:45397", "accept": false, - "reason": "Different failures: one is a position_ids buffer access issue, the other is a Gemma-4 zero3 from_pretrained/config loading problem." + "reason": "Different bugs: position_ids buffer access vs. DeepSpeed ZeRO3 from_pretrained loading failure." }, { "left": "issue:44246", "right": "issue:44496", "accept": false, - "reason": "One is intermittent import slowness; the other is an unrecognized model/config parsing error. Same broad loading area, different bug." + "reason": "Import-time slowdown is unrelated to an unrecognized model/config loading error." }, { "left": "issue:41950", "right": "issue:44220", "accept": false, - "reason": "Different pipelines and code paths: video classification mistakenly uses image processors, while the other is a feature extraction issue." + "reason": "Video-classification processor lookup and audio fbank extraction are different code paths and symptoms." }, { "left": "issue:43531", "right": "issue:45362", "accept": false, - "reason": "Both involve Qwen models, but the symptoms differ: sliding_window handling vs a chat-generation crash on Qwen3.5-35B." + "reason": "Both involve Qwen variants, but the reported failures are different concrete issues in different paths." }, { "left": "issue:42886", "right": "issue:46132", "accept": false, - "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior are unrelated issues." + "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior are unrelated." }, { "left": "issue:36246", "right": "issue:42757", "accept": false, - "reason": "Both are ImportError reports, but for different symbols and packages: Qwen2_5_VLImageProcessor vs is_offline_mode from huggingface_hub." + "reason": "Both are import errors, but for different symbols from different packages and different root causes." }, { "left": "issue:43526", "right": "issue:45003", "accept": false, - "reason": "BeitImageProcessorFast reduce_labels bug is unrelated to unsafe sys.modules access in modeling_utils." + "reason": "BeitImageProcessorFast label reduction and sys.modules access are unrelated bugs." }, { "left": "issue:36246", "right": "issue:44351", "accept": false, - "reason": "Different missing imports in different subsystems; these are not the same underlying failure." + "reason": "Missing Qwen2_5_VLImageProcessor and missing HybridCache are separate import problems." }, { "left": "issue:44336", "right": "issue:44485", "accept": false, - "reason": "ANSI escape code leakage in loading_report is unrelated to the GLM-5 RoPE implementation discussion." + "reason": "ANSI output in loading_report is unrelated to GLM-5 RoPE implementation." }, { "left": "issue:45003", "right": "issue:45561", "accept": false, - "reason": "Unsafe sys.modules access and a pytest-xdist race on captured_info.txt are different concurrency/IO problems." + "reason": "Unsafe sys.modules access and pytest-xdist file race are different failure modes." }, { "left": "issue:43352", "right": "issue:45250", "accept": false, - "reason": "Specific model support error for NemotronHForCausalLM is not the same as the generic Flash Attention 2.0 issue." + "reason": "These only share the Flash Attention 2.0 theme; they do not show the same concrete model/code-path bug." }, { "left": "issue:43673", "right": "issue:44561", "accept": false, - "reason": "Chunked-prefill cache handling and removal of is_torch_fx_available are separate v5 regression reports." + "reason": "Chunked-prefill cache handling and trust_remote_code breakage from a removed helper are unrelated." }, { "left": "issue:44246", "right": "issue:44871", "accept": false, - "reason": "Import performance and Gemma eos_token_id configuration mismatch are unrelated." + "reason": "Import latency and Gemma-3 eos_token_id config mismatch are unrelated." }, { "left": "issue:44242", "right": "issue:44857", "accept": false, - "reason": "Load-balancing loss omission in MoE routing is not the same bug as an AMP/CUDA crash in LwDetrImageLoss." + "reason": "MoE router loss handling and RT-DETR AMP crash are different model-specific bugs." }, { "left": "issue:36246", "right": "issue:43596", "accept": false, - "reason": "Qwen2.5 VLImageProcessor import failure and DeepSpeed/BertModel index error are unrelated." + "reason": "Qwen2_5_VLImageProcessor import failure and BertModel ZeRO3 index error are unrelated." }, { "left": "issue:44162", "right": "issue:45200", "accept": false, - "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaults are different model-specific regressions." + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaulting are different issues." }, { - "left": "issue:45412", - "right": "issue:46123", + "left": "issue:44295", + "right": "issue:44749", "accept": false, - "reason": "Memory not released after RT-DETR deletion is unrelated to MaskGenerationPipeline dropping final partial-batch results." + "reason": "position_ids buffer access and slow filtering after upgrade are not the same underlying defect." }, { - "left": "issue:44295", - "right": "issue:44749", + "left": "issue:45412", + "right": "issue:46123", "accept": false, - "reason": "Position_ids buffer access and post-upgrade filtering slowdown are unrelated." + "reason": "RT-DETR memory retention and mask generation partial-batch dropping are unrelated." }, { "left": "issue:44779", "right": "issue:45292", "accept": false, - "reason": "Deepseek tokenization correctness and resize_token_embeddings/output_embeddings mismatch are different code paths." + "reason": "Tokenizer correctness regression and resize_token_embeddings output sync are different code paths." }, { "left": "issue:43901", "right": "issue:45003", "accept": false, - "reason": "Docs mismatch for return_all_scores is unrelated to sys.modules access in modeling_utils." + "reason": "Docs mismatch for return_all_scores and unsafe sys.modules access are unrelated." }, { "left": "issue:45292", "right": "issue:45584", "accept": false, - "reason": "Embedding resize behavior and Whisper empty transcription after align_special_tokens are unrelated." - }, - { - "left": "issue:45987", - "right": "issue:46082", - "accept": false, - "reason": "StaticCache.get_seq_length return type and LlamaConfig head_dim validation are separate API issues." + "reason": "Embedding resize propagation and Whisper empty transcription failures do not share a concrete root cause." }, { "left": "issue:44496", "right": "issue:45003", "accept": false, - "reason": "Unrecognized model/config loading and unsafe sys.modules access are different failure modes." + "reason": "Unrecognized model config and sys.modules access are separate loading bugs." }, { "left": "issue:45292", "right": "issue:45701", "accept": false, - "reason": "Resize-token-embeddings behavior is unrelated to version-dependent tokenization changes." + "reason": "Resize-token-embeddings behavior and version-dependent tokenization changes are not the same bug." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "Cache sequence-length return type and LlamaConfig head_dim validation are unrelated." }, { "left": "issue:44987", "right": "issue:45459", "accept": false, - "reason": "Physical-intelligence/fast loading failure and hidden protobuf decode errors are different tokenizer/loading bugs." + "reason": "Model-loading failure on 5.1.0 and protobuf error masking are distinct tokenizer/loading issues." }, { "left": "issue:44877", "right": "issue:45375", "accept": false, - "reason": "Strict config rejection for granite_speech is not the same as a missing deepstack_visual_indexes field in Qwen3_5MoeVisionConfig." + "reason": "Strict config rejection for granite_speech and a missing Qwen vision config field are only loosely related by strict config handling, not the same bug." }, { "left": "issue:43232", "right": "issue:44336", "accept": false, - "reason": "Generation kwargs update after sync_gpus and ANSI output in loading_report are unrelated." + "reason": "Generation kwargs sync handling and ANSI codes in loading_report are unrelated." }, { "left": "issue:45375", "right": "issue:45759", "accept": false, - "reason": "Missing config field in Qwen3_5MoeVisionConfig and unwrapping text_config in composite Qwen multimodal configs are distinct issues." + "reason": "Missing config field under strict mode and composite Qwen text_config unwrapping are different config-loading bugs." }, { "left": "issue:43257", "right": "issue:45137", "accept": false, - "reason": "Qwen3 MoE weight conversion with accelerate+deepspeed is unrelated to a DeepSpeed ZeRO3 deque underflow." + "reason": "Qwen3 MOE weight conversion and ZeRO3 deque underflow are different DeepSpeed-related failures." }, { "left": "issue:43262", "right": "issue:44246", "accept": false, - "reason": "Audio chat-template sampling-rate defaulting and import-time slowness do not share a code path." + "reason": "Audio chat-template sampling-rate default and import-time slowdown are unrelated." }, { "left": "issue:44995", "right": "issue:45200", "accept": false, - "reason": "Stale indexer cache crash in GlmMoeDsa and Gemma 4 token-type defaults are unrelated." + "reason": "GlmMoeDsa stale cache on second forward and Gemma 4 token-type defaults are unrelated." } ] }, @@ -11243,14 +11309,13 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ "issue:28282", "issue:30990", - "issue:41628", "issue:42886", "issue:42994", "issue:43065", @@ -11282,6 +11347,7 @@ "issue:44995", "issue:45137", "issue:45229", + "issue:45230", "issue:45278", "issue:45310", "issue:45335", @@ -11302,9 +11368,9 @@ "issue:43475|issue:43540", "issue:42886|issue:45229", "issue:43653|issue:44162", - "issue:44060|issue:44077", "issue:44995|issue:45588", "issue:44466|issue:45335", + "issue:44060|issue:44077", "issue:44368|issue:44977", "issue:45397|issue:45656", "issue:44368|issue:45563", @@ -11329,228 +11395,228 @@ "issue:43475|issue:43643", "issue:46077|issue:46123", "issue:44038|issue:44623", - "issue:41628|issue:44933" + "issue:45230|issue:45397" ], - "prepared_review_unit_hash": "e2562c88e5ef264350a9b2f47963ff243ef651ea9a015dfac1a77b6e270d4caa", + "prepared_review_unit_hash": "97110cb9ac456dd0e7df1328222449434f7f7232d996891eeeda9209e0a49bb8", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 32, - "serialized_chars": 37189, - "estimated_input_tokens": 9298, - "estimated_eval_tokens": 18852 + "serialized_chars": 37133, + "estimated_input_tokens": 9284, + "estimated_eval_tokens": 18824 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is heterogeneous: it mixes unrelated import errors, tokenizer/serialization regressions, generation bugs, and model-specific failures, mostly around Transformers v5. None of the soft pairs look like the same underlying bug/change, so they should all stay separate.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical duplicate emerges; the issues are thematically related but describe different failures in different code paths and models.", + "summary": "These issues are mostly a grab bag of unrelated Transformers regressions across SAM3, Qwen, Gemma, tokenizers, generation, and DeepSpeed. The soft-similarity pairs are thematically related in a few places, but they do not look like the same concrete bug/change, so none should be merged as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set ('many import errors after update from 4.57.0 to 5.5.0') and is the best single representative if one must be chosen, though the cluster is too heterogeneous to treat as one duplicate family.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the broadest umbrella report in the set and best represents the cluster\u2019s overall v5 import/regression theme, even though it is not a duplicate of the others.", + "best_issue_reason": "issue:45278 is the most generally useful representative because it describes a wide regression affecting multiple imports after upgrade, which is closer to a cluster-level symptom than the other narrowly scoped bugs.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43065", "right": "issue:43408", "accept": false, - "reason": "SAM3 pixel decoder dummy layer vs SAM3 video/tracker model-type mismatch; different symptoms and code paths." + "reason": "Both are SAM3-related, but one is about a dummy Conv2d implementation and the other is a model-type/config warning; different bugs." }, { "left": "issue:43475", "right": "issue:43540", "accept": false, - "reason": "Missing fpn_position_embeddings in SAM3 vision output vs video-input ValueError in Qwen3OmniMoe; unrelated models and bugs." + "reason": "Different model families and failure modes: missing encoder output field vs video-input processing error." }, { "left": "issue:42886", "right": "issue:45229", "accept": false, - "reason": "Offline tokenizer cache loading bug vs multi-GPU Gemma4 CUDA OOM; not the same failure." + "reason": "Offline tokenizer cache loading and multi-GPU CUDA OOM are unrelated." }, { "left": "issue:43653", "right": "issue:44162", "accept": false, - "reason": "BigBirdTokenizer special-token registration issue vs ESM2 broken model behavior; different tokenizer/model problems." - }, - { - "left": "issue:44060", - "right": "issue:44077", - "accept": false, - "reason": "Incorrect tied-weights warning in Qwen3-Next vs patchtsmixer post_init compatibility policy; unrelated." + "reason": "Tokenizer special-token registration bug vs a broad ESM2 breakage report; not the same code path." }, { "left": "issue:44995", "right": "issue:45588", "accept": false, - "reason": "Stale indexer cache crash on second forward pass vs flash_attention sink-less AttributeError; different code paths." + "reason": "Stale indexer cache crash and flash-attention s_aux=None AttributeError are distinct failures." }, { "left": "issue:44466", "right": "issue:45335", "accept": false, - "reason": "Device-dependent tied-weight serialization bug vs resize_token_embeddings not updating decoder.embed_tokens; different mechanisms." + "reason": "Both mention tied/embedding behavior, but one is serialization of lm_head.weight and the other is resize_token_embeddings not updating decoder.embed_tokens." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Both concern model/config warnings, but they describe different models and different underlying problems." }, { "left": "issue:44368", "right": "issue:44977", "accept": false, - "reason": "Tie_word_embeddings warning in Qwen3.5 LoRA vs flash-attention generation failure; not the same issue." + "reason": "A tie_word_embeddings warning is not the same as a flash-attention generation failure." }, { "left": "issue:45397", "right": "issue:45656", "accept": false, - "reason": "Gemma4 zero3 from_pretrained failure vs DeepSpeed double optimizer step; separate distributed-training bugs." + "reason": "Gemma4 from_pretrained/zero3 loading and a double optimizer-step issue in DeepSpeed are separate bugs." }, { "left": "issue:44368", "right": "issue:45563", "accept": false, - "reason": "Tie-word-embeddings warning vs stale num_return_sequences warning in paged generate(); both are warnings but not the same bug." + "reason": "One is a configuration warning about tied embeddings; the other is a stale paged-generate warning." }, { "left": "issue:45137", "right": "issue:45600", "accept": false, - "reason": "DeepSpeed ZeRO3 deque underflow vs Sam3LiteText removed-config CI breakage; unrelated failures." + "reason": "DeepSpeed ZeRO3 deque/index error vs broken auto_mappings references to removed Sam3LiteText configs; different subsystems and symptoms." }, { "left": "issue:43673", "right": "issue:44734", "accept": false, - "reason": "Generation cache missing in chunked_prefill vs serve /v1/responses KV-cache indexing crash; different generation layers." + "reason": "Both involve cache/generation, but one is missing GenerationMixin cache in chunked prefill and the other is a serve-side KV-cache indexing crash." }, { "left": "issue:43065", "right": "issue:46123", "accept": false, - "reason": "SAM3 decoder dummy Conv2d vs mask pipeline final-batch bookkeeping bug; unrelated." + "reason": "SAM3 pixel decoder implementation issue vs a MaskGenerationPipeline partial-batch bookkeeping bug; unrelated." }, { "left": "issue:43931", "right": "issue:44368", "accept": false, - "reason": "Qwen3-VL-30B shape mismatch vs Qwen3.5 tied-embeddings warning; different model initialization issues." + "reason": "Qwen3-VL-30B weight-shape mismatch is not the same as a tie_word_embeddings warning." }, { "left": "issue:44933", "right": "issue:45459", "accept": false, - "reason": "Removed image_utils import vs protobuf decode-error handling; both are import-related but not the same bug." + "reason": "Removed image_utils import vs protobuf-related tokenizer error handling; different modules and failure causes." }, { "left": "issue:44869", "right": "issue:45356", "accept": false, - "reason": "Whisper trailing replacement-character decode crash vs Kimi tokenizer regression; different tokenizer logic." + "reason": "Whisper timestamp decode crash and Kimi-K2.5 tokenizer regression are unrelated tokenizer/decoder bugs." }, { "left": "issue:43232", "right": "issue:43673", "accept": false, - "reason": "sync_gpus generation kwargs update bug vs chunked_prefill cache missing; separate generation regressions." + "reason": "Both touch generation internals, but sync_gpus update_kwargs behavior is not the same as missing cache during chunked prefill." }, { "left": "issue:44297", "right": "issue:44623", "accept": false, - "reason": "tokenizer_class mismatch on save_pretrained vs processor.save_pretrained missing files; adjacent serialization area but different defects." + "reason": "save_pretrained bugs, but one is tokenizer_class metadata mismatch and the other is processor missing files." }, { "left": "issue:44361", "right": "issue:44749", "accept": false, - "reason": "MLukeTokenizer AttributeError vs a 10x slowdown in data filtering after upgrade; not the same symptom or path." + "reason": "MLukeTokenizer task AttributeError and a performance regression after upgrade are different problems." }, { "left": "issue:44871", "right": "issue:46132", "accept": false, - "reason": "Gemma-3 eos_token_id inconsistency vs AttentionInterface.register behavior change; unrelated configuration vs API behavior." + "reason": "Gemma eos_token_id config mismatch is unrelated to AttentionInterface.register changing function behavior." }, { "left": "issue:43646", "right": "issue:44162", "accept": false, - "reason": "Generic custom model initialization breakage vs ESM2 broken; one is framework-level, the other model-specific." + "reason": "Generic custom-model initialization breakage and ESM2 breakage are not the same concrete bug." }, { "left": "issue:43673", "right": "issue:45278", "accept": false, - "reason": "Generation cache regression vs broad import errors after upgrade; too different to be duplicates." + "reason": "One is a specific generation cache regression; the other is a broad import-error report after upgrade." }, { "left": "issue:44623", "right": "issue:45656", "accept": false, - "reason": "Missing processor save files vs duplicate optimizer.step under DeepSpeed; different subsystems." + "reason": "Processor save_pretrained missing files and a deepspeed double-step issue are unrelated." }, { "left": "issue:44749", "right": "issue:45656", "accept": false, - "reason": "Performance regression in filtering vs optimizer-step duplication; not the same underlying bug." + "reason": "Performance slowdown after upgrade vs optimizer stepping twice; no shared code-path bug." }, { "left": "issue:44749", "right": "issue:45310", "accept": false, - "reason": "Filtering slowdown after upgrade vs Qwen3.5 MoE from_pretrained error; different behaviors." + "reason": "Chinese performance regression and Qwen3.5 Moe from_pretrained error are different issues." }, { "left": "issue:42994", "right": "issue:43482", "accept": false, - "reason": "Quantized model saving failure vs Qwen2.5-GGUF loading failure in v5; save vs load and different model formats." + "reason": "Quantized model saving failure vs GGUF loading failure are opposite directions and different code paths." }, { "left": "issue:28282", "right": "issue:30990", "accept": false, - "reason": "Missing PyTorch dependency import error vs Sentence Transformers loading hang; different failure modes." + "reason": "PyTorch missing import error vs Sentence Transformers getting stuck loading are not the same bug." }, { "left": "issue:43475", "right": "issue:45381", "accept": false, - "reason": "SAM3 vision encoder attribute error vs Qwen2.5-VL video position-id bug; unrelated multimodal models." + "reason": "Missing fpn_position_embeddings in SAM3 video output and Qwen2.5-VL video position_ids are unrelated video-model bugs." }, { "left": "issue:43475", "right": "issue:43643", "accept": false, - "reason": "SAM3 encoder output attribute missing vs AutoConfig trust_remote_code missing fields; different object-construction bugs." + "reason": "Different model families and different initialization/config issues." }, { "left": "issue:46077", "right": "issue:46123", "accept": false, - "reason": "Gemma4 attention-mask construction issue vs pipeline final partial-batch dropping results; unrelated." + "reason": "Gemma4 attention-mask construction and batch-finalization logic in MaskGenerationPipeline are separate bugs." }, { "left": "issue:44038", "right": "issue:44623", "accept": false, - "reason": "Qwen3-VL-Moe compatibility issue vs processor.save_pretrained missing files; different code paths." + "reason": "Qwen3-VL-Moe from_pretrained error and processor.save_pretrained missing files do not describe the same defect." }, { - "left": "issue:41628", - "right": "issue:44933", + "left": "issue:45230", + "right": "issue:45397", "accept": false, - "reason": "Cannot import AutoImageProcessor vs nonexistent image_utils import; both imports, but not the same missing symbol or breakage." + "reason": "Generic bug report and Gemma4 zero3 from_pretrained failure are not duplicates." } ] }, @@ -11569,8 +11635,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -11588,6 +11654,7 @@ "issue:43232", "issue:43377", "issue:43475", + "issue:43479", "issue:43504", "issue:43525", "issue:43526", @@ -11597,6 +11664,7 @@ "issue:43873", "issue:43976", "issue:44038", + "issue:44079", "issue:44188", "issue:44222", "issue:44242", @@ -11605,6 +11673,7 @@ "issue:44568", "issue:44661", "issue:44792", + "issue:44869", "issue:44933", "issue:44945", "issue:44964", @@ -11613,17 +11682,15 @@ "issue:45103", "issue:45137", "issue:45200", - "issue:45230", "issue:45310", "issue:45375", - "issue:45397", "issue:45478", "issue:45561", "issue:45712", "issue:45753" ], "soft_pairs": [ - "issue:45230|issue:45397", + "issue:41628|issue:44933", "issue:42371|issue:45092", "issue:43475|issue:44661", "issue:42994|issue:43525", @@ -11652,222 +11719,236 @@ "issue:42175|issue:44933", "issue:42898|issue:44568", "issue:43976|issue:44933", - "issue:30990|issue:35141" + "issue:30990|issue:35141", + "issue:43479|issue:44869", + "issue:43479|issue:44079" ], - "prepared_review_unit_hash": "a7e037ee7badb2c01c6401960f7c59757ecf72f326217dd37ca18716810cae48", + "prepared_review_unit_hash": "81796dffe2fd9d9d6fee8e20ecfd570091b0e810facdf3e4a78124bdeaf886c2", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 30, - "serialized_chars": 35908, - "estimated_input_tokens": 8977, - "estimated_eval_tokens": 18210 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37228, + "estimated_input_tokens": 9307, + "estimated_eval_tokens": 18870 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly unrelated Transformers bug reports, but two duplicate groups stand out: the exact-title Qwen3.5 MoE from_pretrained pair (45310/45478) and the exact-title Qwen3ForCausalLM VRAM leak pair (42673/43704). The remaining soft edges mostly look like nearby-but-distinct regressions rather than true duplicates.", - "confidence": 0.79, - "canonical_issue_reason": "Issue 45310 is the best canonical issue anchor because it is the earliest report in the exact-title Qwen3.5 MoE from_pretrained duplicate pair with 45478, and it cleanly represents that specific regression. It also sits at the center of the small Qwen3.5 MoE-related neighborhood better than the later duplicate.", + "summary": "Most pairs are only superficially similar and should stay separate. I tightened the Qwen3.5 triage: 45375/45478 is not a safe duplicate, while the stronger duplicate signal in that cluster is the identical 45310/45478 title pair. Among the provided soft edges, only 42673/43704 is a true duplicate.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45310 is the best issue anchor for the Qwen3.5 MoE from_pretrained regression because it is concrete and its matching duplicate signal is the identical-title report in issue:45478; issue:45375 describes a different missing-field/config serialization problem.", "canonical_pr_reason": null, - "best_issue_reason": "45310 is the strongest global issue candidate because it is a concrete, high-signal bug report with an exact duplicate at 45478. The other items are mostly isolated import, config, or runtime regressions, with 42673/43704 forming the other clear duplicate pair.", + "best_issue_reason": "issue:45310 is the strongest overall issue candidate because it represents the clearest repeated Qwen3.5 MoE regression and is the cleanest anchor for the true duplicate pair with issue:45478.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:45230", - "right": "issue:45397", + "left": "issue:41628", + "right": "issue:44933", "accept": false, - "reason": "Generic placeholder report vs a specific Gemma 4 zero3 bug; not the same underlying issue." + "reason": "Both are import errors, but they involve different missing symbols and different code paths." }, { "left": "issue:42371", "right": "issue:45092", "accept": false, - "reason": "TF32 API guidance issue and an InternVL2 meta-init compatibility bug are unrelated." + "reason": "TF32 configuration behavior and multimodal meta-initialization compatibility are unrelated." }, { "left": "issue:43475", "right": "issue:44661", "accept": false, - "reason": "SAM3 vision encoder attribute error vs tokenizer-mapping model registration failure; different code paths." + "reason": "A missing vision-output attribute is not the same bug as a tokenizer-mapping registration failure." }, { "left": "issue:42994", "right": "issue:43525", "accept": false, - "reason": "Quantized model saving failure and missing pad_token_id on Llama4Config are different bugs." + "reason": "Quantized save failure and missing config attribute are different defects." }, { "left": "issue:45081", "right": "issue:45103", "accept": false, - "reason": "Mistral tokenizer regex crash and auto_docstring annotation crash are unrelated subsystems." + "reason": "Tokenizer regex patching and auto-docstring annotation handling crash in different subsystems." }, { "left": "issue:45375", "right": "issue:45478", "accept": false, - "reason": "Same model family, but one is a missing config field and the other is a from_pretrained failure; not enough evidence of the same bug." + "reason": "Related Qwen3.5 reports, but one is a missing deepstack field while the other is a generic from_pretrained error; not the same concrete bug." }, { "left": "issue:45310", "right": "issue:45375", "accept": false, - "reason": "Both touch Qwen3.5 MoE, but one is a generic from_pretrained error and the other is a specific config-field-dropping issue." + "reason": "Both are Qwen3.5 MoE-related, but the symptoms differ enough that they are not clearly the same bug." }, { "left": "issue:43873", "right": "issue:44945", "accept": false, - "reason": "Quantization/offloading behavior and pipeline-parallel output correctness are different problems." + "reason": "VRAM leakage in dataloader threads and incorrect output under pipeline parallelism are different problems." }, { "left": "issue:41628", "right": "issue:44351", "accept": false, - "reason": "AutoImageProcessor import failure and HybridCache import failure are distinct missing-export issues." + "reason": "They are both imports, but they concern different missing top-level symbols." }, { "left": "issue:43232", "right": "issue:44188", "accept": false, - "reason": "Generation kwargs bug after sync_gpus and torch.compile attention-kernel divergence do not describe the same failure." + "reason": "Generation kwargs syncing and torch.compile attention-kernel branching are unrelated failures." }, { "left": "issue:42673", "right": "issue:43704", "accept": true, - "reason": "Exact same title, same model, same VRAM leak symptom; clear duplicate report." + "reason": "Same Qwen3ForCausalLM VRAM leak report with matching title and symptom." }, { "left": "issue:42831", "right": "issue:44222", "accept": false, - "reason": "FineGrainedFP8 accuracy regression and FP8 save_pretrained failure are different FP8 issues." + "reason": "FP8 accuracy degradation and FP8 save_pretrained failure are distinct issues." }, { "left": "issue:44945", "right": "issue:45753", "accept": false, - "reason": "Pipeline-parallel LLM output corruption and Qwen3_5 image infinite loop are unrelated symptoms." + "reason": "Pipeline-parallel output corruption and a Qwen3_5 image-specific infinite loop are different bugs." }, { "left": "issue:36010", "right": "issue:41628", "accept": false, - "reason": "GenerationMixin import error and AutoImageProcessor import error are separate import regressions." + "reason": "GenerationMixin import failure and AutoImageProcessor import failure are separate regressions." }, { "left": "issue:44222", "right": "issue:45310", "accept": false, - "reason": "FP8 MoE save_pretrained failure is unrelated to the Qwen3.5 MoE from_pretrained error." + "reason": "FP8 save_pretrained for moe and Qwen3.5 MoE from_pretrained error are different code paths." }, { "left": "issue:42994", "right": "issue:43824", "accept": false, - "reason": "Quantized saving bug and Qwen2.5-VL import error are different breakages." + "reason": "Quantized model saving and Qwen2.5-VL import failure are unrelated." }, { "left": "issue:42175", "right": "issue:45712", "accept": false, - "reason": "Backend dependency packaging issue and leftover dummy-class cleanup issue are unrelated." + "reason": "Missing TF backend from extras and leftover dummy classes leaking into the repo are unrelated packaging issues." }, { "left": "issue:43673", "right": "issue:44188", "accept": false, - "reason": "Cache-missing generation bug and attention-kernel divergence under torch.compile are different code paths." + "reason": "Chunked-prefill cache handling and attention-kernel divergence are not the same bug." }, { "left": "issue:43232", "right": "issue:44792", "accept": false, - "reason": "Generation kwargs handling and Janus image-generation test failure are not the same bug." + "reason": "Generation kwargs update and a janus image-generation test failure are different failures." }, { "left": "issue:44038", "right": "issue:45310", "accept": false, - "reason": "Qwen3-VL-Moe compatibility issue and Qwen3.5 MoE from_pretrained error are different model-specific regressions." + "reason": "Both involve Qwen MoE models, but the reported symptoms and model variants differ enough that they are not the same concrete bug." }, { "left": "issue:44964", "right": "issue:45137", "accept": false, - "reason": "Phi-4 multimodal load failure and DeepSpeed ZeRO3 deque error are unrelated." + "reason": "Phi-4 multimodal loading and a DeepSpeed ZeRO3 deque error are unrelated." }, { "left": "issue:44496", "right": "issue:45561", "accept": false, - "reason": "Model config loading error and pytest-xdist file race are separate concerns." + "reason": "Model config recognition failure and pytest-xdist file race are unrelated." }, { "left": "issue:43526", "right": "issue:45561", "accept": false, - "reason": "BeitImageProcessorFast label reduction bug and xdist race on captured_info.txt are unrelated." + "reason": "Label reduction bug and pytest-xdist file race are unrelated." }, { "left": "issue:43504", "right": "issue:45561", "accept": false, - "reason": "BEiT pretrained preset legacy-field issue and xdist race condition are not the same bug." + "reason": "Legacy-field model loading bug and pytest-xdist file race are unrelated." }, { "left": "issue:42222", "right": "issue:45200", "accept": false, - "reason": "Broken vitpose models and Gemma 4 mm_token_type_ids defaulting are different model families and failures." + "reason": "Broken VitPose models and Gemma 4 token-type defaults are different model-specific issues." }, { "left": "issue:43377", "right": "issue:44242", "accept": false, - "reason": "MIMI batching/padding-mask inconsistency and MoE load-balancing loss omission are unrelated behaviors." + "reason": "Padding-mask support in MIMI and missing load-balancing loss when router logits are off are different issues." }, { "left": "issue:42175", "right": "issue:44933", "accept": false, - "reason": "Packaging backend omission and missing image_utils import are different issues." + "reason": "Missing TF backend from extras and nonexistent image_utils import are unrelated." }, { "left": "issue:42898", "right": "issue:44568", "accept": false, - "reason": "Tokenization-spaces behavior change and BOS/EOS insertion regression are separate tokenizer behavior bugs." + "reason": "Tokenization cleanup-space behavior and special-token insertion are different regressions." }, { "left": "issue:43976", "right": "issue:44933", "accept": false, - "reason": "Python version compatibility issue and missing image_utils import are unrelated." + "reason": "Python version compatibility and nonexistent image_utils import are unrelated." }, { "left": "issue:30990", "right": "issue:35141", "accept": false, - "reason": "Sentence Transformers loading hang and embedding reinitialization after resize are different model-loading/training bugs." + "reason": "A loading stall in Sentence Transformers and token-embedding reinitialization are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config initialization and Whisper word-timestamp decoding are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Default config initialization and ModelOutput key assignment are unrelated." } ] }, "evaluator_result": { "accept": false, - "feedback": "The summary is not fully grounded in the packet. It says the only clear duplicate soft edge is the Qwen3ForCausalLM VRAM leak pair, but the packet also contains an exact-title duplicate pair for issue 45310 and 45478 (`[BUG] transformers>=5.4.0, Qwen3.5 Moe from_pretrained error`). The conservative verdicts are mostly fine, but the summary should acknowledge both clear duplicate pairs rather than singling out only one." + "feedback": "The review is not conservative enough on the Qwen3.5 pair. In the packet, issue:45375 is about a missing `deepstack_visual_indexes` field in `Qwen3_5MoeVisionConfig`, while issue:45478 is the generic `Qwen3.5 Moe from_pretrained error`. Those are related, but the evidence does not clearly establish the same concrete bug, so accepting that soft-edge pair overstates the packet. The summary also overstates by saying the only clear duplicate Qwen3.5 pair is represented by 45375/45478; the stronger duplicate signal is the identical 45310/45478 title pair, not 45375/45478." }, "evaluator_used": true, "retried": true, @@ -11883,8 +11964,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -11893,6 +11974,7 @@ "issue:30333", "issue:33290", "issue:34689", + "issue:35141", "issue:36331", "issue:41628", "issue:42757", @@ -11907,7 +11989,6 @@ "issue:43824", "issue:43874", "issue:44060", - "issue:44079", "issue:44188", "issue:44190", "issue:44206", @@ -11916,11 +11997,11 @@ "issue:44297", "issue:44373", "issue:44448", + "issue:44484", "issue:44568", "issue:44623", "issue:44734", "issue:44779", - "issue:44869", "issue:44933", "issue:45071", "issue:45092", @@ -11938,8 +12019,6 @@ "issue:46132" ], "soft_pairs": [ - "issue:43479|issue:44869", - "issue:43479|issue:44079", "issue:44623|issue:45310", "issue:43278|issue:45137", "issue:44188|issue:44373", @@ -11954,9 +12033,9 @@ "issue:45405|issue:45712", "issue:43874|issue:45092", "issue:42898|issue:44448", - "issue:44448|issue:46132", "issue:44933|issue:45712", "issue:44295|issue:45656", + "issue:44448|issue:46132", "issue:44222|issue:44297", "issue:43116|issue:44190", "issue:44297|issue:45656", @@ -11972,246 +12051,248 @@ "issue:45071|issue:46082", "issue:30333|issue:34689", "issue:28282|issue:30333", - "issue:44297|issue:45310" + "issue:44297|issue:45310", + "issue:30333|issue:35141", + "issue:44222|issue:44484" ], - "prepared_review_unit_hash": "ade93e1a6ce02b6814aa6a2eadd41ff9303a9bffe8c0da02f007989cc43c1ce0", + "prepared_review_unit_hash": "52da0c376fc795bcf92b5b5b9590089d3e3ae14f38e46322e5af238663f948f4", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 35, - "serialized_chars": 38210, - "estimated_input_tokens": 9553, - "estimated_eval_tokens": 19362 + "serialized_chars": 38218, + "estimated_input_tokens": 9555, + "estimated_eval_tokens": 19366 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The candidate pairs are mostly unrelated issues that only share broad subsystem keywords (tokenization, configs, deepspeed, imports, etc.). I would reject every soft edge and treat this cluster as heterogeneous rather than a duplicate set.", - "confidence": 0.95, - "canonical_issue_reason": "No true canonical duplicate emerges from these items; if forced to anchor the cluster, issue 45812 is the strongest standalone regression report because it is concrete, user-visible, and broadly scoped.", + "summary": "This set is mostly a loose collection of unrelated Transformers issues, with several soft similarities around tokenizer/v5 regressions but no strong duplicate pairs. The best representative issue is the generic tokenizer regression report, while no PRs are present.", + "confidence": 0.58, + "canonical_issue_reason": "issue:42898 is the broadest, most representative user-facing regression in the set: a concrete v5 behavior change affecting tokenization, which is the main theme shared by several of the softer similarity candidates.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 45812 is the best standalone issue in the set: it describes a clear silent v4\u2192v5 regression with broad impact across Granite models and a concrete failure mode.", + "best_issue_reason": "issue:42898 is the clearest standalone bug report with a specific version boundary and reproducible tokenizer behavior change; it is the most suitable anchor for this cluster.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43479", - "right": "issue:44869", - "accept": false, - "reason": "Different code paths: multimodal config defaults vs Whisper timestamp decoding." - }, - { - "left": "issue:43479", - "right": "issue:44079", - "accept": false, - "reason": "Config initialization bug is unrelated to ModelOutput key assignment." - }, { "left": "issue:44623", "right": "issue:45310", "accept": false, - "reason": "Missing files on save_pretrained is unrelated to Qwen3.5 MoE from_pretrained loading." + "reason": "Both mention save/from_pretrained-style failures, but one is about processor serialization and the other is a Qwen3.5 MoE load error; different code paths and failure modes." }, { "left": "issue:43278", "right": "issue:45137", "accept": false, - "reason": "Embedding dtype drift in eval is unrelated to a DeepSpeed ZeRO3 deque error." + "reason": "BF16/FP32 dtype drift during train/eval is unrelated to the DeepSpeed ZeRO3 deque IndexError." }, { "left": "issue:44188", "right": "issue:44373", "accept": false, - "reason": "Attention kernel divergence is unrelated to a docstring typo." + "reason": "Attention-kernel divergence on torch.compile is unrelated to a wrong docstring for position_ids." }, { "left": "issue:42898", "right": "issue:45812", "accept": false, - "reason": "Both are v5 regressions, but one is tokenization cleanup behavior and the other is wrong token IDs for Granite models." + "reason": "Both are tokenizer regressions in v5, but one is a generic clean_up_tokenization_spaces behavior change and the other is Granite token IDs being wrong; different concrete bugs." }, { "left": "issue:43784", "right": "issue:43824", "accept": false, - "reason": "Sentence-transformers import failure is unrelated to missing Qwen2_5_VL class export." + "reason": "Importing sentence-transformers with missing nn is a dependency/import issue, not the missing Qwen2_5_VL model class." }, { "left": "issue:30333", "right": "issue:33290", "accept": false, - "reason": "MLflow reporting on failed jobs is unrelated to Adafactor OOMs in DeepSpeed." + "reason": "MLflow reporting on failed training jobs is unrelated to an OOM in DeepSpeed with Adafactor." }, { "left": "issue:44734", "right": "issue:45103", "accept": false, - "reason": "Serve KV-cache indexing crash is unrelated to auto-docstring annotation handling." + "reason": "KV-cache continuation tensor indexing in serve is unrelated to the auto_docstring crash under future annotations." }, { "left": "issue:43723", "right": "issue:44568", "accept": false, - "reason": "Both are tokenizer regressions, but they affect different APIs and failure modes." + "reason": "Both are tokenizer regressions, but one is tokenizer loading in v5 and the other is add_special_tokens/BOS-EOS handling for a specific tokenizer; not the same underlying bug." }, { "left": "issue:42898", "right": "issue:45356", "accept": false, - "reason": "Different tokenizer regressions: clean_up_tokenization_spaces behavior vs Kimi-K2.5 codec handling." + "reason": "Both are tokenizer-related regressions, but one concerns generic tokenization-space cleanup while the other is Kimi-K2.5 codec handling and a warning message." }, { "left": "issue:42898", "right": "issue:44779", "accept": false, - "reason": "Different models and symptoms; these are separate tokenizer regressions." + "reason": "Deepseek tokenization correctness and clean_up_tokenization_spaces are distinct tokenizer bugs with different symptoms and likely fixes." }, { "left": "issue:44060", "right": "issue:45125", "accept": false, - "reason": "Tied-weights warning bug is unrelated to missing tensor-parallel plan support." + "reason": "A wrong tied-weights warning for Qwen3-Next is unrelated to a missing _tp_plan for Qwen3_5Moe tensor parallelism." }, { "left": "issue:45405", "right": "issue:45712", "accept": false, - "reason": "Dependency version bump issue is unrelated to dummy PT object leakage." + "reason": "PEFT version pinning and leftover dummy classes are unrelated maintenance issues." }, { "left": "issue:43874", "right": "issue:45092", "accept": false, - "reason": "Image processor API break is unrelated to remote-code meta-init incompatibility." + "reason": "A missing image-patch count method in GLM46V is not the same bug as old InternVL2 checkpoints failing meta initialization." }, { - "left": "issue:42898", - "right": "issue:44448", + "left": "issue:44933", + "right": "issue:45712", "accept": false, - "reason": "General v5 tokenization behavior change is unrelated to Pegasus output regression." + "reason": "A nonexistent import from image_utils is unrelated to dummy PT object cleanup." }, { - "left": "issue:44448", - "right": "issue:46132", + "left": "issue:42898", + "right": "issue:44448", "accept": false, - "reason": "Pegasus output difference is unrelated to AttentionInterface.register behavior change." + "reason": "Pegasus output differences between v4 and v5 are a model-generation regression, not the same as generic tokenization-space cleanup." }, { - "left": "issue:44933", - "right": "issue:45712", + "left": "issue:44295", + "right": "issue:45656", "accept": false, - "reason": "Missing image_utils import is unrelated to dummy class cleanup." + "reason": "Registering position_ids as a buffer causing read errors is unrelated to optimizer.step being called twice with DeepSpeed." }, { - "left": "issue:44295", - "right": "issue:45656", + "left": "issue:44448", + "right": "issue:46132", "accept": false, - "reason": "Position_ids buffer access bug is unrelated to optimizer stepping twice in DeepSpeed." + "reason": "Different output for Pegasus and AttentionInterface.register changing callback behavior are unrelated concerns." }, { "left": "issue:44222", "right": "issue:44297", "accept": false, - "reason": "FP8 save_pretrained bug is unrelated to tokenizer config class mismatch." + "reason": "FP8 save_pretrained for MoE models and tokenizer.save_pretrained metadata mismatch are different serialization bugs." }, { "left": "issue:43116", "right": "issue:44190", "accept": false, - "reason": "Text classification example bug is unrelated to image-classification local dataset loading." + "reason": "Empty multi-label classification results in an example script is unrelated to loading a local dataset in another training script." }, { "left": "issue:44297", "right": "issue:45656", "accept": false, - "reason": "Tokenizer save_pretrained metadata bug is unrelated to DeepSpeed optimizer stepping." + "reason": "Tokenizer save_pretrained class metadata mismatch is unrelated to DeepSpeed optimizer double-step behavior." }, { "left": "issue:43479", "right": "issue:43644", "accept": false, - "reason": "Default multimodal config initialization is unrelated to non-persistent buffer corruption." + "reason": "Incorrect default multimodal config initialization is unrelated to non-persistent buffers being filled with junk." }, { "left": "issue:44190", "right": "issue:46082", "accept": false, - "reason": "Dataset loading failure is unrelated to LlamaConfig head_dim validation." + "reason": "Local dataset loading in an example script is unrelated to LlamaConfig head_dim validation." }, { "left": "issue:42831", "right": "issue:44623", "accept": false, - "reason": "FineGrainedFP8 accuracy issue is unrelated to processor save_pretrained missing files." + "reason": "FineGrainedFP8 accuracy issues are not the same as missing files from processor.save_pretrained." }, { "left": "issue:44222", "right": "issue:44623", "accept": false, - "reason": "MoE FP8 save_pretrained is unrelated to processor save_pretrained missing files." + "reason": "Both involve save_pretrained, but one is an FP8 MoE save issue and the other is missing processor files; different artifacts and failure modes." }, { "left": "issue:45245", "right": "issue:46132", "accept": false, - "reason": "Category limit error is unrelated to AttentionInterface.register behavior." + "reason": "Category cardinality limits are unrelated to AttentionInterface.register behavior changes." }, { "left": "issue:44206", "right": "issue:45103", "accept": false, - "reason": "Feature extractor center-arg crash is unrelated to auto-docstring future-annotations crash." + "reason": "A LasrFeatureExtractor argument regression is unrelated to an auto_docstring annotation crash." }, { "left": "issue:30333", "right": "issue:36331", "accept": false, - "reason": "Failed-job MLflow reporting is unrelated to CustomTrainer compute_loss signature mismatch." + "reason": "MLflow reporting for failed jobs and CustomTrainer.compute_loss signature changes are separate training-stack bugs." }, { "left": "issue:41628", "right": "issue:42757", "accept": false, - "reason": "Transformers AutoImageProcessor import failure is unrelated to huggingface_hub offline-mode import failure." + "reason": "Missing AutoImageProcessor import and missing is_offline_mode from huggingface_hub are different import breakages." }, { "left": "issue:22355", "right": "issue:28282", "accept": false, - "reason": "onnx import error is unrelated to missing PyTorch for AutoModel." + "reason": "Missing transformers.onnx and missing PyTorch for AutoModel are unrelated dependency/import issues." }, { "left": "issue:45071", "right": "issue:46082", "accept": false, - "reason": "PretrainedConfig type-checking break is unrelated to LlamaConfig head_dim validation." + "reason": "PretrainedConfig type checking regression is unrelated to LlamaConfig head_dim validation." }, { "left": "issue:30333", "right": "issue:34689", "accept": false, - "reason": "MLflow reporting on failed jobs is unrelated to Llama 3.2 model loading regression." + "reason": "MLflow failure reporting and Llama 3.2 vision model loading are unrelated." }, { "left": "issue:28282", "right": "issue:30333", "accept": false, - "reason": "Missing PyTorch import error is unrelated to MLflow reporting on failed training jobs." + "reason": "PyTorch-not-found ImportError and MLflow reporting on failed training are unrelated." }, { "left": "issue:44297", "right": "issue:45310", "accept": false, - "reason": "Tokenizer class metadata bug is unrelated to Qwen3.5 MoE from_pretrained loading failure." + "reason": "Tokenizer class metadata mismatch in save_pretrained is unrelated to Qwen3.5 MoE from_pretrained loading failure." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "Training-job MLflow reporting and embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 MoE save_pretrained and the max_shard_size default question are not the same bug." } ] }, @@ -12230,14 +12311,14 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ - "issue:30333", + "issue:22355", + "issue:28218", "issue:30990", - "issue:35141", "issue:38175", "issue:39401", "issue:41628", @@ -12246,13 +12327,15 @@ "issue:42886", "issue:42907", "issue:42994", + "issue:43066", "issue:43232", "issue:43479", "issue:43723", + "issue:43756", "issue:43761", "issue:43824", "issue:44038", - "issue:44222", + "issue:44060", "issue:44261", "issue:44263", "issue:44295", @@ -12260,7 +12343,6 @@ "issue:44373", "issue:44442", "issue:44483", - "issue:44484", "issue:44485", "issue:44514", "issue:44568", @@ -12284,14 +12366,12 @@ "issue:46082" ], "soft_pairs": [ - "issue:44222|issue:44484", - "issue:30333|issue:35141", "issue:43723|issue:44977", "issue:43232|issue:45834", "issue:44263|issue:45092", "issue:43761|issue:46082", - "issue:44483|issue:45406", "issue:44373|issue:44485", + "issue:44483|issue:45406", "issue:44263|issue:44360", "issue:44295|issue:44623", "issue:44514|issue:45103", @@ -12310,218 +12390,221 @@ "issue:42831|issue:44038", "issue:39401|issue:44568", "issue:42757|issue:45278", - "issue:42886|issue:42994", "issue:43232|issue:45593", - "issue:42831|issue:45753" + "issue:42886|issue:42994", + "issue:42831|issue:45753", + "issue:43756|issue:44060", + "issue:42886|issue:43066", + "issue:22355|issue:28218" ], - "prepared_review_unit_hash": "3d9015c64d21af1a45c546428cb9148c97775d3d18a1e0a556a359b997f6166f", + "prepared_review_unit_hash": "ce4ec132738576a0cc51da25d7412123a518cc12d61c46434caedad4bdcbc33a", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 29, - "serialized_chars": 35509, - "estimated_input_tokens": 8878, - "estimated_eval_tokens": 18012 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36333, + "estimated_input_tokens": 9084, + "estimated_eval_tokens": 18424 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a grab-bag of unrelated Transformers/HF bugs. There are a few loose themes around v5 tokenizer/import/load regressions and save/load edge cases, but none of the soft pairs are the same concrete defect, so all soft edges are rejected.", - "confidence": 0.88, - "canonical_issue_reason": "No true cluster-wide duplicate exists; issue 43723 is the broadest representative because it describes a generic v5 tokenizer-loading regression.", + "summary": "Mostly unrelated issue reports with a loose concentration of tokenizer/import regressions; none of the soft pairs are strong enough to treat as true duplicates from titles alone.", + "confidence": 0.42, + "canonical_issue_reason": "issue:43723 is the broadest and most representative of the repeated v5 tokenizer-loading regression theme, so it is the best single issue anchor in this mixed set.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43723 is the most generally scoped and actionable report among the loading/import regressions, so it is the best single representative if one must be chosen.", + "best_issue_reason": "issue:43723 best fits the recurring tokenizer-loading/import failure pattern, but the cluster is too heterogeneous for a strong canonical choice.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44222", - "right": "issue:44484", - "accept": false, - "reason": "Both mention save_pretrained, but one is FP8/MoE saving and the other is a generic shard-size question; different problems." - }, - { - "left": "issue:30333", - "right": "issue:35141", - "accept": false, - "reason": "MLflow job reporting and embedding reinitialization are unrelated bugs in different code paths." - }, { "left": "issue:43723", "right": "issue:44977", "accept": false, - "reason": "Tokenizer-loading regression vs flash-attention generation failure; different failure modes and subsystems." + "reason": "One is a tokenizer-loading regression; the other is a Qwen3.5 generation/flash-attention problem. Different code paths and symptoms." }, { "left": "issue:43232", "right": "issue:45834", "accept": false, - "reason": "Generation kwargs sync issue and OCR index error are unrelated model/runtime bugs." + "reason": "Generation kwargs handling after sync_gpus is unrelated to Kosmos2.5 long OCR input indexing." }, { "left": "issue:44263", "right": "issue:45092", "accept": false, - "reason": "DSA indexer torch.split issue is unrelated to remote-code meta-init incompatibility." + "reason": "torch.split/indexer behavior and InternVL2 meta-init compatibility are unrelated bugs." }, { "left": "issue:43761", "right": "issue:46082", "accept": false, - "reason": "Hidden_states regression in CLIP vision vs LlamaConfig head_dim validation; not the same bug." + "reason": "CLIPVisionModel hidden_states regression and LlamaConfig head_dim validation are different subsystems and failures." }, { - "left": "issue:44483", - "right": "issue:45406", + "left": "issue:44373", + "right": "issue:44485", "accept": false, - "reason": "Chat/completions request handling and Gemma4Processor missing _tokenizer are different code-path failures." + "reason": "A docstring error for position_ids is not the same as a GLM-5 RoPE implementation bug." }, { - "left": "issue:44373", - "right": "issue:44485", + "left": "issue:44483", + "right": "issue:45406", "accept": false, - "reason": "Docstring mismatch for position_ids is not the same as a RoPE implementation bug." + "reason": "Chat/completions request validation and Gemma4Processor missing _tokenizer are distinct problems." }, { "left": "issue:44263", "right": "issue:44360", "accept": false, - "reason": "Two different DSA-indexer complaints with different symptoms and fixes." + "reason": "Both touch the DSA indexer, but one is torch.split return values while the other is a missing ReLU; not the same bug." }, { "left": "issue:44295", "right": "issue:44623", "accept": false, - "reason": "position_ids buffer access error and processor save_pretrained missing files are unrelated." + "reason": "Reading position_ids after buffer registration and processor.save_pretrained missing files are different failure modes." }, { "left": "issue:44514", "right": "issue:45103", "accept": false, - "reason": "Processor batching crash and auto-docstring crash are entirely different issues." + "reason": "Qwen2_5_VLProcessor batch padding crash is unrelated to the auto_docstring future-annotations AttributeError." }, { "left": "issue:43723", "right": "issue:44871", "accept": false, - "reason": "Generic tokenizer-loading regression vs eos_token_id configuration mismatch; different concrete defects." + "reason": "Tokenizer-loading regression vs mismatched eos_token_id configuration; related to tokenizers, but not the same underlying bug." }, { "left": "issue:45092", "right": "issue:45357", "accept": false, - "reason": "Meta-init incompatibility for old InternVL2 checkpoints is not the same as incorrect visual encoder key saving." + "reason": "InternVL2 meta-init incompatibility and Qwen3.5 save_pretrained key corruption are separate model-specific save/load issues." }, { "left": "issue:43723", "right": "issue:44991", "accept": false, - "reason": "Both are tokenizer load failures, but for different models and likely different causes; too broad to merge." + "reason": "Both are tokenizer-load failures in v5, but one is generic and the other is model-specific; insufficient evidence of one shared root cause." }, { "left": "issue:30990", "right": "issue:38175", "accept": false, - "reason": "Sentence Transformers loading hang and SigLIP zero-probabilities are unrelated." + "reason": "Sentence Transformers loading hang and SigLIP2 zero-probabilities are unrelated." }, { "left": "issue:42907", "right": "issue:46050", "accept": false, - "reason": "Both concern quantized save/load, but different quantization methods and failure modes." + "reason": "Different quantization save/load problems with different methods and symptoms; not clearly the same bug." }, { "left": "issue:44829", "right": "issue:46002", "accept": false, - "reason": "Flash-attention training degeneration and NaNs at checkout/init are different training stability bugs." + "reason": "Flash-attention training degeneration and NaNs after checkpoint/init are not the same code-path issue." }, { "left": "issue:43723", "right": "issue:44442", "accept": false, - "reason": "Tokenizer load regression in v5 and a FastSpeech2ConformerTokenizer import/load failure are separate concrete issues." + "reason": "Both involve AutoTokenizer loading, but the metadata is too different to conclude a single duplicate bug." }, { "left": "issue:44261", "right": "issue:45092", "accept": false, - "reason": "Missing rms_norm_eps in MLA layernorm and InternVL2 meta-init incompatibility are unrelated." + "reason": "MLA layernorm epsilon precision and InternVL2 meta-init are unrelated." }, { "left": "issue:41628", "right": "issue:43824", "accept": false, - "reason": "Different missing imports: AutoImageProcessor vs Qwen2_5_VLForConditionalGeneration." + "reason": "Both are import errors, but for different symbols and likely different root causes; not a true duplicate." }, { "left": "issue:42831", "right": "issue:45310", "accept": false, - "reason": "FineGrainedFP8 accuracy issue and Qwen3.5 MoE from_pretrained error are not the same defect." + "reason": "FineGrainedFP8 accuracy and Qwen3.5 from_pretrained error are unrelated." }, { "left": "issue:43479", "right": "issue:44913", "accept": false, - "reason": "Config default initialization bug and config reload persistence bug affect different models/fields." - }, - { - "left": "issue:42757", - "right": "issue:45278", - "accept": false, - "reason": "Specific missing hub symbol vs broad import-error report after upgrade; related theme, not one concrete bug." + "reason": "Config default initialization on None and rotary_pct not persisting on reload are different config bugs." }, { - "left": "issue:42886", - "right": "issue:42994", - "accept": false, - "reason": "Offline tokenizer cache loading and quantized model saving failures are unrelated." - }, - { - "left": "issue:43232", - "right": "issue:45593", + "left": "issue:44263", + "right": "issue:44373", "accept": false, - "reason": "Generation state update bug and D-FINE auxiliary-loss behavior are different code paths." + "reason": "Indexing/torch.split behavior and a docstring typo are not the same underlying issue." }, { "left": "issue:42831", "right": "issue:44038", "accept": false, - "reason": "FP8 accuracy regression is not the same as the Qwen3-VL-Moe bug." + "reason": "Accuracy with FineGrainedFP8 and a Qwen3-VL-Moe bug are not the same change or failure." }, { "left": "issue:39401", "right": "issue:44568", "accept": false, - "reason": "Offset_mapping bug for Qwen3 and missing BOS/EOS insertion for mdeberta-v3 are different tokenizer behaviors." + "reason": "Offset mapping wrong for Qwen3 and add_special_tokens missing BOS/EOS are distinct tokenizer bugs." }, { "left": "issue:42757", "right": "issue:45278", "accept": false, - "reason": "Even though both are import-related, they are not the same underlying missing-symbol defect." + "reason": "A missing huggingface_hub symbol and broad import errors after upgrade are related only at a very high level; not enough for duplication." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation-state handling and D-FINE auxiliary-loss behavior are unrelated." }, { "left": "issue:42886", "right": "issue:42994", "accept": false, - "reason": "Cache/offline tokenizer loading and quantized save_pretrained failure do not share the same root cause." + "reason": "Offline tokenizer cache loading and quantized model saving are different workflows and bugs." }, { "left": "issue:42831", "right": "issue:45753", "accept": false, - "reason": "FP8 accuracy regression and Qwen3_5 infinite-loop-on-image are different problems." + "reason": "FineGrainedFP8 accuracy and Qwen3_5 infinite-loop image behavior are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Smollm3 RoPE-layer dropping and Qwen3-Next tied-weights warning are different model/config issues." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline cache loading failure and wrong decoder type in v5 are different tokenizer issues." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Missing transformers.onnx module and extra-space tokenization behavior are unrelated." } ] }, @@ -12540,40 +12623,39 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ - "issue:22355", - "issue:28218", "issue:29127", "issue:30990", "issue:33357", + "issue:36296", "issue:39401", "issue:41628", "issue:42371", - "issue:42886", "issue:42907", - "issue:43066", "issue:43116", "issue:43122", "issue:43232", "issue:43408", + "issue:43452", "issue:43475", "issue:43576", "issue:43646", - "issue:43756", "issue:43784", + "issue:43824", "issue:43937", "issue:43976", "issue:44038", - "issue:44060", "issue:44117", + "issue:44222", "issue:44261", "issue:44263", "issue:44295", "issue:44297", + "issue:44351", "issue:44360", "issue:44373", "issue:44485", @@ -12584,6 +12666,7 @@ "issue:44671", "issue:44704", "issue:44743", + "issue:44779", "issue:44829", "issue:44945", "issue:44964", @@ -12595,10 +12678,7 @@ "issue:46132" ], "soft_pairs": [ - "issue:43756|issue:44060", - "issue:42886|issue:43066", "issue:43122|issue:44521", - "issue:22355|issue:28218", "issue:43576|issue:43784", "issue:44671|issue:46132", "issue:44117|issue:44704", @@ -12607,8 +12687,8 @@ "issue:42371|issue:44263", "issue:44704|issue:45290", "issue:43232|issue:44743", - "issue:44295|issue:44297", "issue:41628|issue:45278", + "issue:44295|issue:44297", "issue:42907|issue:44945", "issue:43122|issue:44514", "issue:43976|issue:44945", @@ -12623,210 +12703,234 @@ "issue:43116|issue:45200", "issue:42907|issue:44964", "issue:44038|issue:44297", - "issue:29127|issue:30990" + "issue:29127|issue:30990", + "issue:43122|issue:44779", + "issue:36296|issue:43116", + "issue:43824|issue:44351", + "issue:44038|issue:44295", + "issue:42907|issue:43452", + "issue:44038|issue:44222" ], - "prepared_review_unit_hash": "e3370503b0461e6b30e5a64dbe7b9008d3b83e9988ca0fd762d61169cc2bbf9d", + "prepared_review_unit_hash": "e228d15e2d0065acabe271c1bbaa4f26f0acb70dc14c6a261c4ad50b7f1fb170", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 29, - "serialized_chars": 35986, - "estimated_input_tokens": 8997, - "estimated_eval_tokens": 18250 + "soft_pair_count": 32, + "serialized_chars": 36845, + "estimated_input_tokens": 9212, + "estimated_eval_tokens": 18680 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items form a loose cluster of Transformers regressions and tokenizer/loader issues, but the soft edges are mostly thematic overlaps rather than true duplicates. I would not merge any of the suggested pairs.", - "confidence": 0.78, - "canonical_issue_reason": "issue:43122 is the strongest representative of the tokenizer-regression subgroup: it has a clear cross-version failure, a concrete repro angle, and the closest semantic ties to the other tokenizer/chat-template reports.", + "summary": "The cluster is heterogeneous: most pairs are only loosely related by broad Transformers subsystems or version-regression language, not the same underlying bug/change. No soft-edge pair looks safe to merge as a duplicate.", + "confidence": 0.9, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:43122 is the best standalone anchor because it describes a specific, reproducible tokenizer behavior change and sits nearest to the other tokenizer-related soft matches.", + "best_issue_reason": "issue:45278 is the broadest regression-style report in the set (many import errors after upgrading to v5.5.0) and is the best loose representative if a single issue must anchor the cluster, but it does not match most of the other items closely.", "best_pr_reason": null, "soft_edge_verdicts": [ { - "left": "issue:43756", - "right": "issue:44060", + "left": "issue:43122", + "right": "issue:44521", "accept": false, - "reason": "Different model-specific bugs: RoPE layer dropping vs an incorrect tied-weights warning." + "reason": "Both involve tokenization/template behavior, but one is a version-to-version tokenization change and the other is multimodal assistant-mask generation; different bugs." }, { - "left": "issue:42886", - "right": "issue:43066", + "left": "issue:43576", + "right": "issue:43784", "accept": false, - "reason": "Both are tokenizer-adjacent, but one is offline cache loading and the other is a decoder-type issue." + "reason": "A broken `transformers env` command and an import-time `nn` NameError are unrelated failure modes." }, { - "left": "issue:43122", - "right": "issue:44521", + "left": "issue:44671", + "right": "issue:46132", "accept": false, - "reason": "Cross-version tokenization drift is not the same as multimodal assistant mask generation." - }, - { - "left": "issue:22355", - "right": "issue:28218", - "accept": false, - "reason": "Unrelated failures: missing onnx module import vs extra whitespace after added tokens." - }, - { - "left": "issue:43576", - "right": "issue:43784", - "accept": false, - "reason": "Different surfaces: broken env command vs import-time NameError in sentence-transformers." - }, - { - "left": "issue:44671", - "right": "issue:46132", - "accept": false, - "reason": "Masked LM prediction regression and AttentionInterface.register behavior change are distinct bugs." + "reason": "Incorrect CamemBERT MLM predictions and `AttentionInterface.register` changing function behavior are distinct code paths." }, { "left": "issue:44117", "right": "issue:44704", "accept": false, - "reason": "Tokenizer mapping fallback handling is separate from AutoProcessor kwarg forwarding." + "reason": "Tokenizer mapping fallback handling vs `AutoProcessor.from_pretrained` kwargs forwarding are separate API issues." }, { "left": "issue:43646", "right": "issue:46002", "accept": false, - "reason": "Custom model initialization breakage is not the same as NaNs after checkout/init." + "reason": "Custom model initialization breakage and NaNs in classification heads are different underlying problems." }, { "left": "issue:30990", "right": "issue:33357", "accept": false, - "reason": "A loading hang and a MacOS bus error are different failure modes." + "reason": "Loading hangs for Sentence Transformers and a MacOS bus error with a CLIP model are not the same bug." }, { "left": "issue:42371", "right": "issue:44263", "accept": false, - "reason": "TF32 settings guidance is unrelated to the torch.split return-value issue." + "reason": "TF32 API guidance and a `torch.split()` return-value issue in GlmMoeDsaIndexer are unrelated." }, { "left": "issue:44704", "right": "issue:45290", "accept": false, - "reason": "Different code paths: processor kwargs passthrough vs chat-template crash with tool calls." + "reason": "`cached_file` kwarg forwarding and chat-template crashes with tool-call messages are different processor paths." }, { "left": "issue:43232", "right": "issue:44743", "accept": false, - "reason": "Generation kwargs after sync_gpus and recurrent-state reset in modular_qwen3_5 are different regressions." + "reason": "Both touch generation/cache logic, but one is `_update_model_kwargs_for_generation` after `sync_gpus` and the other is recurrent-state reset in a Qwen module; not the same defect." }, { - "left": "issue:44295", - "right": "issue:44297", + "left": "issue:41628", + "right": "issue:45278", "accept": false, - "reason": "Buffer access error for position_ids is unrelated to tokenizer_class metadata mismatch on save." + "reason": "A single missing `AutoImageProcessor` import is narrower than the broad multi-import regression in v5.5.0." }, { - "left": "issue:41628", - "right": "issue:45278", + "left": "issue:44295", + "right": "issue:44297", "accept": false, - "reason": "A single import error is not the same concrete bug as broader post-upgrade import regressions." + "reason": "Reading `position_ids` after buffer registration and saving tokenizer class metadata are unrelated." }, { "left": "issue:42907", "right": "issue:44945", "accept": false, - "reason": "Saving dequantized models and pipeline-parallel output corruption are separate issues." + "reason": "Dequantized model saving and incorrect output under pipeline parallelism are different functional areas." }, { "left": "issue:43122", "right": "issue:44514", "accept": false, - "reason": "Tokenizer behavior change and batched apply_chat_template padding crash are different bugs." + "reason": "Both mention tokenization/chat templates, but the first is a general version tokenization shift and the second is a batched multimodal padding crash." }, { "left": "issue:43976", "right": "issue:44945", "accept": false, - "reason": "Python version compatibility and pipeline parallelism correctness are unrelated." + "reason": "Python version compatibility and pipeline-parallel inference correctness are unrelated." }, { "left": "issue:39401", "right": "issue:44488", "accept": false, - "reason": "Wrong offset_mapping and failure to load a specific model are not the same bug." + "reason": "Qwen3 offset mapping errors and a model load failure for `cjvt/sleng-bert` do not describe the same bug." }, { "left": "issue:44261", "right": "issue:44373", "accept": false, - "reason": "A numerical precision/config issue is unrelated to a wrong docstring." + "reason": "A numerical precision/config issue and a wrong docstring are not duplicates." }, { "left": "issue:43784", "right": "issue:45600", "accept": false, - "reason": "Import NameError and auto_mappings CI breakage from removed configs are different causes." + "reason": "Import-time `nn` failure in sentence-transformers and stale `auto_mappings.py` config references are different import regressions." }, { "left": "issue:42907", "right": "issue:43937", "accept": false, - "reason": "Dequantized-save failure and GenerationConfig validation error are separate problems." + "reason": "Saving dequantized models and invalid `GenerationConfig` errors are separate problems." }, { "left": "issue:43408", "right": "issue:43475", "accept": false, - "reason": "Both are SAM3-related, but one is a type-mismatch warning and the other is a missing attribute error." + "reason": "The first is a model-type mismatch warning; the second is a missing attribute on a SAM3 output object." }, { "left": "issue:44360", "right": "issue:44485", "accept": false, - "reason": "DSA indexer activation omission and GLM-5 RoPE implementation discussion are not the same issue." + "reason": "A missing ReLU in the DSA indexer and GLM-5 RoPE implementation issues are unrelated." }, { "left": "issue:39401", "right": "issue:44521", "accept": false, - "reason": "Tokenizer offset mapping and multimodal assistant_mask behavior are different bug reports." + "reason": "Offset-mapping correctness and multimodal assistant masks are different tokenizer/processor bugs." }, { "left": "issue:44625", "right": "issue:44829", "accept": false, - "reason": "num_labels propagation and flash_attention_3 training degeneration are unrelated." + "reason": "`num_labels` propagation for Qwen3.5 and flash-attention-3 degenerate training are separate model/training issues." }, { "left": "issue:43116", "right": "issue:45200", "accept": false, - "reason": "Example-script multi-label behavior and Gemma 4 token-type defaults are different failures." + "reason": "Empty multi-label outputs in an example script and missing default `mm_token_type_ids` are not the same failure." }, { "left": "issue:42907", "right": "issue:44964", "accept": false, - "reason": "Model save failure and a model loading failure for Phi-4 multimodal are distinct." + "reason": "Dequantized save failures and inability to load a multimodal Phi-4 model are different code paths." }, { "left": "issue:44038", "right": "issue:44297", "accept": false, - "reason": "Qwen3-VL-Moe bug and tokenizer save metadata mismatch do not share the same underlying cause." + "reason": "Qwen3-VL-Moe behavior and tokenizer save metadata mismatch are unrelated." }, { "left": "issue:29127", "right": "issue:30990", "accept": false, - "reason": "Clarity of an error message is not the same as a loading hang." + "reason": "LayoutLMv3 box-validation messaging and a loading hang in Sentence Transformers are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different symptoms (`offset_mapping` vs general incorrect tokenization)." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training bug and example-script multi-label output failure are different areas." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Missing Qwen2.5-VL export and missing HybridCache export are separate import errors." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe model behavior and `position_ids` buffer-read issues are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Saving dequantized models and `gguf_file`-related loading failures are different bugs." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "A Qwen3-VL-Moe regression and FP8 save_pretrained behavior for MoE are not the same underlying issue." } ] }, @@ -12845,43 +12949,41 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ "issue:22355", "issue:29127", "issue:29159", - "issue:36296", + "issue:33453", "issue:38617", + "issue:39401", "issue:42222", "issue:42371", "issue:42673", "issue:42757", - "issue:42907", "issue:43064", "issue:43066", "issue:43116", "issue:43122", "issue:43278", "issue:43441", - "issue:43452", "issue:43493", "issue:43611", "issue:43673", "issue:43784", - "issue:43824", "issue:43825", + "issue:43827", "issue:43874", "issue:43937", "issue:43976", "issue:43992", "issue:44038", "issue:44075", - "issue:44222", - "issue:44295", "issue:44351", + "issue:44380", "issue:44485", "issue:44704", "issue:44779", @@ -12897,15 +12999,10 @@ "issue:45478", "issue:45563", "issue:45753", - "issue:46002" + "issue:46002", + "issue:46077" ], "soft_pairs": [ - "issue:43122|issue:44779", - "issue:42907|issue:43452", - "issue:36296|issue:43116", - "issue:43824|issue:44351", - "issue:44038|issue:44295", - "issue:44038|issue:44222", "issue:45216|issue:45478", "issue:42757|issue:44351", "issue:43278|issue:44945", @@ -12930,222 +13027,207 @@ "issue:44038|issue:45478", "issue:43116|issue:43825", "issue:45083|issue:45446", - "issue:43937|issue:45563" + "issue:43937|issue:45563", + "issue:44380|issue:46077", + "issue:33453|issue:39401", + "issue:43827|issue:45446" ], - "prepared_review_unit_hash": "818c5bace63307f93db3ec8f12bcaedb34354f3a6624134667c993aaafb0db8a", + "prepared_review_unit_hash": "ab96bd57220bbb15f12737054938a1d10b9f9d833384787b1d3814418ff382a4", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 31, - "serialized_chars": 36539, - "estimated_input_tokens": 9135, - "estimated_eval_tokens": 18526 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35097, + "estimated_input_tokens": 8775, + "estimated_eval_tokens": 17806 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a loose similarity group of unrelated Transformers bug reports; the only mild commonality is version-regression / tokenizer-related complaints, but there is no true duplicate core.", - "confidence": 0.88, - "canonical_issue_reason": "Issue 43122 is the closest representative of the recurring tokenizer-regression theme, but the cluster is too heterogeneous to treat as a single duplicate set.", + "summary": "The cluster is mostly heterogeneous: a few tokenizer/import/regression themes recur, but the soft pairs are not the same concrete bug/change. No PRs are present. The best issue representative is the broad post-upgrade import-error report.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45278 is the closest umbrella report among the set: it captures a broad post-upgrade import-error regression, making it the most representative issue in an otherwise mixed cluster.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43122 is the best single representative by breadth and topical centrality (tokenization behavior change across versions), though it does not canonically cover the rest of the cluster.", + "best_issue_reason": "issue:45278 is the most broadly representative and can plausibly subsume multiple import-error reports, even though it is not a true duplicate of the specific symbol-level failures.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43122", - "right": "issue:44779", - "accept": false, - "reason": "Both are tokenizer regressions, but 43122 is broad version-to-version tokenization drift while 44779 is Deepseek-specific incorrect output; not the same bug." - }, - { - "left": "issue:42907", - "right": "issue:43452", - "accept": false, - "reason": "Different problems: dequantized save_pretrained failure vs AutoTokenizer/AutoModel loading with gguf_file." - }, - { - "left": "issue:36296", - "right": "issue:43116", - "accept": false, - "reason": "Tensor-parallel training bug and empty multi-label example-script output are unrelated code paths." - }, - { - "left": "issue:43824", - "right": "issue:44351", - "accept": false, - "reason": "Separate import errors for different symbols/models: Qwen2.5-VL class vs HybridCache." - }, - { - "left": "issue:44038", - "right": "issue:44295", - "accept": false, - "reason": "Different symptoms and subsystems: Qwen3-VL-Moe issue vs position_ids buffer access error." - }, - { - "left": "issue:44038", - "right": "issue:44222", - "accept": false, - "reason": "Both touch MoE models, but one is a Qwen3-VL-Moe runtime bug and the other is FP8 save_pretrained behavior; not the same concrete fix." - }, { "left": "issue:45216", "right": "issue:45478", "accept": false, - "reason": "Both concern Qwen3.5 save/load regressions, but one is save_pretrained checkpoint correctness and the other is from_pretrained loading failure." + "reason": "Same model family, but one is a save_pretrained checkpoint-key regression and the other is a from_pretrained load error; related area, not the same concrete bug." }, { "left": "issue:42757", "right": "issue:44351", "accept": false, - "reason": "Two unrelated import failures for different names from different modules." + "reason": "Both are import errors, but for different symbols from different packages/modules; no shared code-path bug is evident." }, { "left": "issue:43278", "right": "issue:44945", "accept": false, - "reason": "BF16/FP32 dtype drift in eval is unrelated to incorrect output under pipeline parallelism." + "reason": "Both touch training/eval behavior, but one is a dtype shift and the other is incorrect output under pipeline parallelism; too different to duplicate." }, { "left": "issue:43784", "right": "issue:45446", "accept": false, - "reason": "Import-time NameError in sentence-transformers versus a PyTorch version guard in flex_attention; different bugs." + "reason": "Unrelated import/version problems: sentence-transformers import NameError versus a flex_attention version check bug." }, { "left": "issue:43122", "right": "issue:44977", "accept": false, - "reason": "Tokenizer version regression vs Qwen3.5 flash-attention generation failure; not the same issue." + "reason": "Both are v5 regressions involving tokenization/generation, but the concrete failures differ: tokenizer output drift vs flash-attention generation failure." }, { "left": "issue:22355", "right": "issue:29159", "accept": false, - "reason": "Missing transformers.onnx module and slow/fast tokenizer inconsistency are unrelated." + "reason": "Module import failure for transformers.onnx is unrelated to slow/fast tokenizer inconsistency." }, { "left": "issue:43493", "right": "issue:45753", "accept": false, - "reason": "Different models and failure modes: SigLIP2 implementation discrepancy vs Qwen3_5 image-specific infinite loop." + "reason": "Different models and failure modes: SigLIP2 implementation discrepancy versus an infinite loop on a specific image." }, { "left": "issue:43066", "right": "issue:44779", "accept": false, - "reason": "Both are tokenizer-related, but one is decoder type handling in v5 and the other is a Deepseek tokenization regression." + "reason": "Both are tokenizer-related regressions, but one is about decoder type selection in v5 and the other is Deepseek output correctness; not the same bug." }, { "left": "issue:43992", "right": "issue:44704", "accept": false, - "reason": "UMT5Encoder weight loading and AutoProcessor cached_file kwargs are separate loading bugs." + "reason": "Two separate from_pretrained-related bugs in different classes; one misses a weight key, the other drops kwargs to cached_file." }, { "left": "issue:38617", "right": "issue:45278", "accept": false, - "reason": "Both are import-related, but one is a single missing symbol and the other is a broad upgrade breakage report." + "reason": "45278 is a broad import-error umbrella, but 38617 is a specific missing symbol in configuration_utils; not clearly the same underlying defect." }, { "left": "issue:42673", "right": "issue:45072", "accept": false, - "reason": "VRAM leak in multi-threaded dataloading and bfloat16 dtype mismatches are unrelated training/inference issues." + "reason": "VRAM leak in Qwen3ForCausalLM threads and bfloat16 dtype mismatches are unrelated." }, { "left": "issue:43937", "right": "issue:44485", "accept": false, - "reason": "Invalid GenerationConfig for GLM-5 and a RoPE implementation discussion are not the same bug." + "reason": "Both mention GLM-5, but one is invalid GenerationConfig and the other is a RoPE implementation discussion; different issues." }, { "left": "issue:22355", "right": "issue:29127", "accept": false, - "reason": "Module import failure and LayoutLMv3 box-validation messaging are unrelated." + "reason": "Missing transformers.onnx module and a LayoutLMv3 error-message clarity issue are not the same bug." }, { "left": "issue:43611", "right": "issue:46002", "accept": false, - "reason": "Base-model-prefix loading regression and NaNs in classification heads are different failures." + "reason": "Model loading via base_model_prefix and NaNs in classification heads are separate failure modes." }, { "left": "issue:43976", "right": "issue:45446", "accept": false, - "reason": "Python version support mismatch and a flex_attention import/version check are not the same issue." + "reason": "Python version support breakage and a PyTorch version-check import issue are unrelated." }, { "left": "issue:43116", "right": "issue:43493", "accept": false, - "reason": "Example-script multi-label output bug and SigLIP2 implementation discrepancy are unrelated." + "reason": "Different subsystems entirely: run_classification multi-label output vs SigLIP2 model discrepancy." }, { "left": "issue:42222", "right": "issue:43441", "accept": false, - "reason": "Broken vitpose models and Ministral-3 FlashAttention failure are different model-specific bugs." + "reason": "VitPose import breakage and Ministral-3 FlashAttention failure are unrelated model-specific bugs." }, { "left": "issue:44075", "right": "issue:44936", "accept": false, - "reason": "Optimizer SGD args not used and trainer.evaluate() failing after train are separate training issues." + "reason": "Optimizer SGD arguments being ignored is not the same as trainer.evaluate failing after train." }, { "left": "issue:42371", "right": "issue:43874", "accept": false, - "reason": "TF32 control warning and missing get_number_of_image_patches method are unrelated." + "reason": "TF32 API guidance and a missing image-processor helper method are unrelated." }, { "left": "issue:43064", "right": "issue:43673", "accept": false, - "reason": "FSDP2/PEFT optimizer-state corruption and missing GenerationMixin cache in chunked_prefill are separate code paths." + "reason": "Optimizer-state corruption under FSDP2/PEFT is unrelated to generation cache handling in chunked_prefill." }, { "left": "issue:45357", "right": "issue:45478", "accept": false, - "reason": "Both involve Qwen3.5 regressions, but one is save_pretrained visual-encoder keys and the other is from_pretrained loading; not mergeable as one fix." + "reason": "Both concern Qwen3.5 save/load behavior, but one is about incorrect saved visual encoder keys and the other is a from_pretrained error; related, but not clearly the same concrete bug." }, { "left": "issue:44038", "right": "issue:45478", "accept": false, - "reason": "Different Qwen models and different failure points; not the same concrete bug." + "reason": "Same broad Qwen3 family, but different model variants and failure descriptions; not enough to treat as one bug." }, { "left": "issue:43116", "right": "issue:43825", "accept": false, - "reason": "Classification example-script output and pipeline translation-task support messaging are unrelated." + "reason": "A classification script bug and a pipeline() translation-task error message issue are unrelated." }, { "left": "issue:45083", "right": "issue:45446", "accept": false, - "reason": "Feature-length helper behavior and a PyTorch version check bug have no direct overlap." + "reason": "Qwen3 Omni-MoE feature-length helper behavior and a flex_attention import/version check are unrelated." }, { "left": "issue:43937", "right": "issue:45563", "accept": false, - "reason": "Invalid generation config and stale num_return_sequences warning are different generation-layer issues." + "reason": "GenerationConfig validation failure and a stale num_return_sequences warning are different generate-path issues." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "Both concern attention behavior, but GPT-2 backend scaling and Gemma4 bidirectional masking are distinct code paths and fixes." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Tokenizer loading regression and Qwen3 offset_mapping correctness are related only at a high level; not the same bug." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() and a PyTorch version-check import problem are unrelated." } ] }, @@ -13164,12 +13246,11 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ - "issue:33453", "issue:38617", "issue:39401", "issue:41628", @@ -13178,21 +13259,23 @@ "issue:42371", "issue:42617", "issue:42673", + "issue:42831", + "issue:43066", + "issue:43122", "issue:43408", "issue:43421", "issue:43450", "issue:43577", "issue:43582", "issue:43742", - "issue:43827", "issue:43906", "issue:43937", "issue:43986", + "issue:44075", "issue:44117", "issue:44220", "issue:44261", "issue:44291", - "issue:44380", "issue:44410", "issue:44466", "issue:44483", @@ -13214,13 +13297,10 @@ "issue:45542", "issue:45561", "issue:45588", - "issue:46002", - "issue:46077" + "issue:45656", + "issue:46002" ], "soft_pairs": [ - "issue:44380|issue:46077", - "issue:33453|issue:39401", - "issue:43827|issue:45446", "issue:44117|issue:45588", "issue:39401|issue:44779", "issue:42673|issue:43906", @@ -13235,221 +13315,224 @@ "issue:44410|issue:45310", "issue:44829|issue:45216", "issue:39401|issue:43421", - "issue:44484|issue:45310", "issue:42673|issue:43582", + "issue:44484|issue:45310", "issue:42617|issue:45341", - "issue:42371|issue:44261", "issue:44933|issue:45278", + "issue:42371|issue:44261", "issue:42673|issue:43577", "issue:42222|issue:45542", "issue:43408|issue:43742", "issue:38617|issue:44220", "issue:43937|issue:44496", "issue:42175|issue:42222", - "issue:44483|issue:45464" + "issue:44483|issue:45464", + "issue:44075|issue:45656", + "issue:43066|issue:43122", + "issue:42371|issue:42831" ], - "prepared_review_unit_hash": "b3323ad41bdecd69795d8f98f1cc52712e0542d33c9167eb7a2c5e9d365dad59", + "prepared_review_unit_hash": "223ed6cc879294fb89be4664e7ae783b5d66bc53be6b2b0f669015c45acc081b", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 47, - "item_count": 47, + "node_count": 48, + "item_count": 48, "soft_pair_count": 29, - "serialized_chars": 35772, - "estimated_input_tokens": 8943, - "estimated_eval_tokens": 18142 + "serialized_chars": 36230, + "estimated_input_tokens": 9058, + "estimated_eval_tokens": 18372 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is noisy and mostly groups superficially similar Transformers issues; the only broadly representative item is a generic tokenizer-loading regression. I found no PRs, and none of the soft pairs look like true duplicates of the same concrete bug.", - "confidence": 0.32, - "canonical_issue_reason": "issue:33453 is the broadest and most central title in the set (\u201cRegression in tokenizer loading\u201d), making it the best umbrella for the tokenizer-focused portion of the cluster.", + "summary": "The cluster is very heterogeneous: most soft matches are superficial similarities across unrelated Transformers issues (tokenizers, model loading, imports, optimizer behavior, testing, and API regressions). I did not find any pair that clearly looks like the same underlying bug or change.", + "confidence": 0.93, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "issue:33453 is the best global issue fit because it is generic enough to cover the recurring tokenizer-loading/thread in the cluster, while the rest are mostly narrower model- or API-specific failures.", + "best_issue_reason": "issue:45278 is the closest single representative because it captures a broad import-regression theme, but the set is too mixed to treat it as a true canonical duplicate hub.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44380", - "right": "issue:46077", - "accept": false, - "reason": "Different model/backends and different failure modes: GPT-2 attention scaling vs Gemma4 bidirectional mask construction." - }, - { - "left": "issue:33453", - "right": "issue:39401", - "accept": false, - "reason": "Both involve tokenizers, but one is loading regression and the other is incorrect offset mapping; not the same concrete bug." - }, - { - "left": "issue:43827", - "right": "issue:45446", - "accept": false, - "reason": "Docs/pipeline removal issue is unrelated to a PyTorch version check in flex_attention." - }, { "left": "issue:44117", "right": "issue:45588", "accept": false, - "reason": "Tokenizer mapping None bug and flash_attention sink-less AttributeError are unrelated code paths." + "reason": "TOKENIZER_MAPPING_NAMES returning None and a flash_attention s_aux=None crash are different code paths and failure modes." }, { "left": "issue:39401", "right": "issue:44779", "accept": false, - "reason": "Different tokenizer regressions affecting different models and behaviors; no clear shared bug." + "reason": "Both are tokenizer regressions, but they concern different models and different symptoms (offset mapping vs incorrect generation/tokenization behavior)." }, { "left": "issue:42673", "right": "issue:43906", "accept": false, - "reason": "The second is only an isolated reproduction reference; insufficient evidence it is the same VRAM leak bug." + "reason": "VRAM leak in multi-threaded dataloaders is unrelated to an isolated reproduction of another issue." }, { "left": "issue:43937", "right": "issue:44977", "accept": false, - "reason": "GenerationConfig validation failure and flash-attention generation issues are distinct problems." + "reason": "GenerationConfig invalid and flash-attention generation failure are distinct bugs with different apparent causes." }, { "left": "issue:44466", "right": "issue:46002", "accept": false, - "reason": "Serialization of tied weights and NaNs in classification heads are different failures." + "reason": "lm_head.weight serialization inconsistency is not the same as NaNs in classification heads during checkout/init." }, { "left": "issue:44291", "right": "issue:45071", "accept": false, - "reason": "init_empty_weights TypeError and PretrainedConfig type-checking regression are separate compatibility issues." + "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig type checking are separate regression points." }, { "left": "issue:43450", "right": "issue:43986", "accept": false, - "reason": "Batched video processor shape bug is unrelated to an AutoProcessor import crash without torchvision." + "reason": "Batched video processor shape bug and missing torchvision AutoProcessor crash are different issues." }, { "left": "issue:45405", "right": "issue:45446", "accept": false, - "reason": "Dependency pin/version-release problem is unrelated to a flex_attention import version check." + "reason": "A pinned PEFT version not yet released and a PyTorch version check bug are unrelated dependency problems." }, { "left": "issue:45341", "right": "issue:45561", "accept": false, - "reason": "Both touch testing_utils, but one is a generic bug report and the other is a specific xdist race on captured_info.txt." + "reason": "A generic testing_utils bug is not the same as the captured_info.txt race under pytest-xdist." }, { "left": "issue:44117", "right": "issue:45081", "accept": false, - "reason": "Tokenizer mapping fallback bug and Mistral regex patch crash are different tokenizer paths." + "reason": "Tokenizer mapping None and Mistral regex patch backend_tokenizer AttributeError are different tokenizer failures." }, { "left": "issue:38617", "right": "issue:41628", "accept": false, - "reason": "Different import errors for different symbols; no evidence of a shared root cause." + "reason": "Different missing imports in different APIs; no shared underlying defect is evident." }, { "left": "issue:44410", "right": "issue:45310", "accept": false, - "reason": "Different model families and distinct missing/failed loading behavior." + "reason": "Missing attn projections in qwen3next and Qwen3.5 MoE from_pretrained failure are different model-specific bugs." }, { "left": "issue:44829", "right": "issue:45216", "accept": false, - "reason": "FlashAttention training degeneration and save_pretrained checkpoint serialization are unrelated." + "reason": "Flash-attention 3 training degeneration and save_pretrained checkpoint corruption are not the same concrete bug." }, { "left": "issue:39401", "right": "issue:43421", "accept": false, - "reason": "Offset mapping bug and runtime post-processor update bug are different tokenizer issues." + "reason": "Offset mapping wrong and post-processor not updating on runtime special-token changes are related to tokenizers but not the same problem." }, { - "left": "issue:44484", - "right": "issue:45310", + "left": "issue:42673", + "right": "issue:43582", "accept": false, - "reason": "Saving shard size behavior and Qwen3.5 MoE from_pretrained error are unrelated." + "reason": "A VRAM leak in dataloader threads is unrelated to an Apple Silicon caching_allocator_warmup TypeError." }, { - "left": "issue:42673", - "right": "issue:43582", + "left": "issue:44484", + "right": "issue:45310", "accept": false, - "reason": "VRAM leak in multithreaded loading and Apple Silicon TypeError in allocator warmup are different bugs." + "reason": "A max_shard_size question is unrelated to a Qwen3.5 MoE from_pretrained error." }, { "left": "issue:42617", "right": "issue:45341", "accept": false, - "reason": "3d_parallel execution failure and a testing_utils bug do not share the same code path." + "reason": "3d_parallel.py execution failure and a testing_utils bug are different issues." }, { - "left": "issue:42371", - "right": "issue:44261", + "left": "issue:44933", + "right": "issue:45278", "accept": false, - "reason": "TF32 behavior settings issue and MLA rms_norm_eps precision issue are unrelated." + "reason": "Both mention import problems, but one is a specific missing import and the other is a broad multi-import regression; not enough evidence they are the same bug." }, { - "left": "issue:44933", - "right": "issue:45278", + "left": "issue:42371", + "right": "issue:44261", "accept": false, - "reason": "A specific missing import and a broad report of many import errors after upgrade are too different to merge as one bug." + "reason": "TF32 API guidance and MLA rms_norm_eps precision error are unrelated." }, { "left": "issue:42673", "right": "issue:43577", "accept": false, - "reason": "VRAM leak in Qwen3ForCausalLM and BLIP2 dtype loading regression are unrelated." + "reason": "VRAM leakage and Blip2 dtype remaining float32 are different bugs." }, { "left": "issue:42222", "right": "issue:45542", "accept": false, - "reason": "VitPose model breakage and missing TensorFlow backend detection with tensorboard are unrelated." + "reason": "Broken vitpose models and a tensorboard/TensorFlow backend error are unrelated." }, { "left": "issue:43408", "right": "issue:43742", "accept": false, - "reason": "sam3_video/sam3_tracker mismatch and MobileLLM key error are different loading issues." + "reason": "SAM3 model-type warning and MobileLLM key error are distinct loading issues." }, { "left": "issue:38617", "right": "issue:44220", "accept": false, - "reason": "ImportError for configuration_utils and an audio feature extraction issue are unrelated." + "reason": "layer_type_validation import failure and _torch_extract_fbank_features() issue are not the same code-path problem." }, { "left": "issue:43937", "right": "issue:44496", "accept": false, - "reason": "Invalid GenerationConfig and unrecognized model/config are distinct model-loading problems." + "reason": "Invalid GenerationConfig and unrecognized model/model_type missing are different failure classes." }, { "left": "issue:42175", "right": "issue:42222", "accept": false, - "reason": "Backend packaging dependency issue and broken VitPose models are not the same bug." + "reason": "TensorFlow backend packaging and vitpose model breakage are unrelated." }, { "left": "issue:44483", "right": "issue:45464", "accept": false, - "reason": "Generic /v1/chat/completions API rejection and Qwen3.5 streaming failure are not the same concrete failure." + "reason": "A /v1/chat/completions API incompatibility and Qwen3.5 streaming inference failure are not clearly the same bug." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "Optimizer args ignored and optimizer step called twice with deepspeed are different optimizer bugs." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Wrong tokenizer decoder type and tokenization changes from 4.57.3 to 5.0 are related but not the same concrete defect." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 behavior settings and FineGrainedFP8 accuracy problems are unrelated." } ] }, @@ -13468,35 +13551,34 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ "issue:38617", "issue:41628", "issue:42222", - "issue:42371", "issue:42757", - "issue:42831", "issue:43012", "issue:43064", - "issue:43066", "issue:43097", - "issue:43122", "issue:43335", "issue:43408", "issue:43441", "issue:43452", + "issue:43502", "issue:43531", "issue:43577", "issue:43673", + "issue:43824", + "issue:43873", "issue:43874", "issue:43931", "issue:43986", + "issue:43992", "issue:44038", "issue:44062", - "issue:44075", "issue:44117", "issue:44261", "issue:44263", @@ -13507,10 +13589,12 @@ "issue:44484", "issue:44485", "issue:44493", + "issue:44704", "issue:44829", "issue:44987", "issue:45003", "issue:45071", + "issue:45103", "issue:45341", "issue:45372", "issue:45375", @@ -13518,14 +13602,10 @@ "issue:45478", "issue:45542", "issue:45588", - "issue:45656", "issue:46050", "issue:46123" ], "soft_pairs": [ - "issue:44075|issue:45656", - "issue:43066|issue:43122", - "issue:42371|issue:42831", "issue:43012|issue:43577", "issue:43673|issue:45341", "issue:44410|issue:45478", @@ -13559,267 +13639,284 @@ "issue:38617|issue:44351", "issue:44062|issue:44117", "issue:44410|issue:45375", - "issue:44038|issue:44410" + "issue:44038|issue:44410", + "issue:43502|issue:43992", + "issue:44704|issue:45103", + "issue:43097|issue:43873", + "issue:43986|issue:44117", + "issue:43824|issue:43986" ], - "prepared_review_unit_hash": "a35fd83d24ff4f592d50a707ad6bdf74d26873f7cf17e64cf063de6d8dfdd86e", + "prepared_review_unit_hash": "5664ec889ede3dfc1bf447ad1cbadd465b6c14769b328776690fc5e0c56d7d3d", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 37, - "serialized_chars": 38747, - "estimated_input_tokens": 9687, - "estimated_eval_tokens": 19630 + "soft_pair_count": 39, + "serialized_chars": 39486, + "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 heterogeneous and mostly contains distinct import errors, model-specific regressions, and unrelated v5 behavior changes. No soft edge looks like a true duplicate pair, so I would not merge any of the suggested links.", - "confidence": 0.37, - "canonical_issue_reason": "No exact duplicate stands out across the set; if one issue must anchor the cluster, issue 43097 is the broadest v5 regression report with explicit API removal and inbound references.", + "summary": "The soft edges are mostly superficial title-level similarities (imports, loading, config, flash-attention, Qwen/GLM families), but they point to different concrete failure modes and should not be merged as duplicates.", + "confidence": 0.79, + "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "Issue 43097 is the best representative of the cluster because it is a concrete, user-facing v5 breakage with broader relevance than the more model-specific reports.", + "best_issue_reason": "issue:43502 is the strongest standalone issue in this batch: it states a concrete, user-visible invariant violation (local_files_only still triggering API requests) with a clear reproduction target and broad impact.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44075", - "right": "issue:45656", - "accept": false, - "reason": "Both involve optimizer behavior, but one is SGD args being ignored and the other is double optimizer stepping with DeepSpeed; different bugs and code paths." - }, - { - "left": "issue:43066", - "right": "issue:43122", - "accept": false, - "reason": "Both are tokenizer-v5 regressions, but one is about decoder type selection and the other about changed tokenization output; not the same defect." - }, - { - "left": "issue:42371", - "right": "issue:42831", - "accept": false, - "reason": "TF32 API guidance and FineGrainedFP8 accuracy loss are unrelated precision topics with different failure modes." - }, { "left": "issue:43012", "right": "issue:43577", "accept": false, - "reason": "A PyTorch warning during bfloat16 compilation is not the same as Blip2 dtype fields staying float32 on load." + "reason": "Both mention dtype/precision, but one is a PyTorch warning during compilation and the other is an incorrect float32 load state in BLIP2; different code paths and symptoms." }, { "left": "issue:43673", "right": "issue:45341", "accept": false, - "reason": "Generation cache loss in v5.0.0 and a testing_utils bug are unrelated subsystems." + "reason": "Completely unrelated areas: generation cache/chunked prefill versus a testing utility bug." }, { "left": "issue:44410", "right": "issue:45478", "accept": false, - "reason": "One is missing projections in qwen3next layer 0, the other is a Qwen3.5 MoE from_pretrained load failure; different model families and symptoms." + "reason": "Different models and failures: missing projections in qwen3next versus a from_pretrained error for Qwen3.5 MoE." }, { "left": "issue:44484", "right": "issue:45478", "accept": false, - "reason": "Save_pretrained shard-size defaults are unrelated to the Qwen3.5 MoE loading error." + "reason": "Unrelated: a save_pretrained shard-size question versus a model loading error." }, { "left": "issue:43097", "right": "issue:44493", "accept": false, - "reason": "API removal for tie_embeddings_and_encoder_decoder is unrelated to unexpected position-id keys during model loading." + "reason": "Both are version/regression style reports, but one is about a removed config field and the other about unexpected position-id keys; not the same bug." }, { "left": "issue:44263", "right": "issue:44485", "accept": false, - "reason": "torch.split return handling in a GLM indexer is a different bug than the GLM-5 RoPE implementation issue." + "reason": "Both are GLM-related, but they describe different concrete faults in different subsystems (torch.split behavior vs RoPE implementation)." }, { "left": "issue:41628", "right": "issue:45071", "accept": false, - "reason": "Missing AutoImageProcessor import and PretrainedConfig type-checking breakage are distinct regressions." + "reason": "Different surfaces entirely: an import failure for AutoImageProcessor versus a PretrainedConfig type-checking regression." }, { "left": "issue:43874", "right": "issue:44261", "accept": false, - "reason": "A missing image-patch helper in an image processor is unrelated to MLA layernorm epsilon precision handling." + "reason": "Different bug classes: a missing image-processor method versus a numerical precision/config issue in MLA q_a_layernorm." }, { "left": "issue:43012", "right": "issue:43408", "accept": false, - "reason": "A bfloat16 compilation warning is not the same as a model-type mismatch warning between sam3_video and sam3_tracker." + "reason": "One is a precision warning; the other is a model-type mismatch warning. Similar in being warnings, but not the same underlying issue." }, { "left": "issue:42222", "right": "issue:45459", "accept": false, - "reason": "VitPose model breakage and tokenizer protobuf-error masking are unrelated." + "reason": "Unrelated: vitpose model breakage versus tokenizer protobuf error handling." }, { "left": "issue:45071", "right": "issue:45542", "accept": false, - "reason": "Type checking in PretrainedConfig and the undefined tf backend when only tensorboard is installed are different issues." + "reason": "Unrelated regression types: config type checking versus tensorboard/TensorFlow backend detection." }, { "left": "issue:43441", "right": "issue:44829", "accept": false, - "reason": "Both mention FlashAttention, but they affect different models and fail in different ways: load-time crash versus degenerate training." + "reason": "Both involve flash attention, but one is a model-specific failure and the other is degenerate training under flash_attention_3; not the same code-path problem." }, { "left": "issue:43931", "right": "issue:44410", "accept": false, - "reason": "Qwen3-VL weight-shape mismatch and qwen3next missing projections are different model-loading problems." + "reason": "Different model families and failure modes: weight-shape mismatch versus missing attention projections." }, { "left": "issue:43673", "right": "issue:46123", "accept": false, - "reason": "Missing GenerationMixin cache in chunked prefill is unrelated to MaskGenerationPipeline dropping the final partial batch." + "reason": "Generation cache missing in chunked prefill versus a pipeline batching flag bug; different code paths." }, { "left": "issue:43931", "right": "issue:44038", "accept": false, - "reason": "A Qwen3-VL weight mismatch and a Qwen3-VL-Moe bug are distinct load failures." + "reason": "Both are Qwen3-VL related, but one is a shape-mismatch load failure and the other is a Qwen3-VL-Moe issue in transformers v5; too different to merge." }, { "left": "issue:43097", "right": "issue:43452", "accept": false, - "reason": "Removing tie_embeddings_and_encoder_decoder and gguf_file breaking tokenizer/model loading are separate regression paths." + "reason": "Different loading problems: removed tie_embeddings handling versus gguf_file breaking tokenizer/model from_pretrained." }, { "left": "issue:45003", "right": "issue:45341", "accept": false, - "reason": "Unsafe sys.modules access in modeling_utils is unrelated to a bug in testing_utils." + "reason": "Unrelated: sys.modules access in modeling_utils versus a testing_utils bug." }, { "left": "issue:43335", "right": "issue:45071", "accept": false, - "reason": "SwitchTransformers sparse-layer creation and PretrainedConfig type checking are unrelated." + "reason": "Different areas and failure modes: SwitchTransformers config construction versus config type checking." }, { "left": "issue:43986", "right": "issue:45341", "accept": false, - "reason": "Video-model loading without torchvision and a testing_utils bug do not describe the same underlying defect." + "reason": "Unrelated: video-model loading without torchvision versus a testing utility bug." }, { "left": "issue:44117", "right": "issue:44987", "accept": false, - "reason": "TOKENIZER_MAPPING_NAMES returning None is a lower-level mapping issue, while loading physical-intelligence/fast is a broader model-load failure; not enough to treat as one bug." + "reason": "Both are loading-related, but one is a tokenizer mapping assumption and the other is a model-specific transformers>=5.1.0 failure; not the same bug." }, { "left": "issue:43408", "right": "issue:44368", "accept": false, - "reason": "A SAM3 model-type warning and a tie_word_embeddings warning on Qwen3.5 are unrelated config warnings." + "reason": "Different models and issues: SAM3 model-type warning versus a tie_word_embeddings warning in Qwen3.5 LoRA fine-tuning." }, { "left": "issue:43874", "right": "issue:44263", "accept": false, - "reason": "Image patch counting and GLM Moe indexer split behavior are unrelated code paths." + "reason": "Both are GLM-adjacent, but one is an image-processor method omission and the other is a torch.split return-value issue; not the same defect." }, { "left": "issue:43986", "right": "issue:44484", "accept": false, - "reason": "Video AutoProcessor crash and save_pretrained shard sizing have no causal overlap." + "reason": "No overlap in bug shape: torchvision-dependent processor loading versus save_pretrained shard sizing." }, { "left": "issue:43986", "right": "issue:45372", "accept": false, - "reason": "Missing torchvision and missing ReasoningEffort import are separate dependency/import problems." + "reason": "Both are processor/loading failures, but the root causes differ completely (missing torchvision versus missing ReasoningEffort import)." }, { "left": "issue:43531", "right": "issue:45341", "accept": false, - "reason": "Qwen3-MoE sliding_window behavior and a testing_utils bug are unrelated." + "reason": "Unrelated: Qwen3-MoE sliding-window behavior versus a testing utility bug." }, { "left": "issue:38617", "right": "issue:42757", "accept": false, - "reason": "Both are ImportErrors, but for different symbols from different packages and unrelated fix paths." + "reason": "Both are import errors, but they involve different missing symbols from different packages and are not the same underlying breakage." }, { "left": "issue:44261", "right": "issue:44485", "accept": false, - "reason": "MLA q_a_layernorm epsilon handling and GLM-5 RoPE implementation are different model-specific issues." + "reason": "Both mention GLM/MLA-related math, but one is an epsilon/config precision issue and the other is a RoPE implementation bug." }, { "left": "issue:43064", "right": "issue:44315", "accept": false, - "reason": "Wrong optimizer states on nonzero ranks with FSDP2/PEFT is not the same as Liger Kernel not being applied under model_init." + "reason": "Different training-stack problems: wrong optimizer state placement under FSDP2/PEFT versus Liger Kernel not being applied with model_init." }, { "left": "issue:43097", "right": "issue:46050", "accept": false, - "reason": "A removed config option in v5 and quantized model save/load breakage are distinct bugs." + "reason": "Different persistence/load regressions: removed tie_embeddings_and_encoder_decoder versus broken quantized save/load flow." }, { "left": "issue:43097", "right": "issue:45588", "accept": false, - "reason": "API removal for tie_embeddings_and_encoder_decoder does not match the flash_attention.py s_aux=None crash." + "reason": "Unrelated code paths: config-field removal versus a flash_attention crash on s_aux=None." }, { "left": "issue:38617", "right": "issue:44351", "accept": false, - "reason": "Different missing imports from different modules; no shared underlying defect." + "reason": "Both are import-name errors, but they are different missing APIs and likely different release breakages." }, { "left": "issue:44062", "right": "issue:44117", "accept": false, - "reason": "AddedToken constructor argument duplication and tokenizer mapping returning None are unrelated tokenizer issues." + "reason": "Tokenizer construction error versus tokenizer mapping returning None; different concrete problems." }, { "left": "issue:44410", "right": "issue:45375", "accept": false, - "reason": "Missing qwen3next projections and a missing deepstack_visual_indexes config field are distinct model-definition problems." + "reason": "Different model/config fields and different failures; not the same bug." }, { "left": "issue:44038", "right": "issue:44410", "accept": false, - "reason": "Qwen3-VL-Moe from_pretrained failure and qwen3next layer projection omission are different model families and different failures." - } - ] - }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, - "error_kind": null, - "error_message": null + "reason": "Related model family, but distinct failures: a Qwen3-VL-Moe loading issue versus missing projections in qwen3next." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "Different load-time issues: unwanted network access versus missing embed_tokens.weight in UMT5Encoder." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "Both involve kwargs handling, but one is AutoProcessor forwarding to cached_file and the other is an auto_docstring AttributeError under future annotations; not the same bug." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Different areas: config-field removal versus offloading behavior under quantization." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Different failure modes: missing torchvision during video-model loading versus TOKENIZER_MAPPING_NAMES returning None." + }, + { + "left": "issue:43824", + "right": "issue:43986", + "accept": false, + "reason": "Different models and causes: missing Qwen2_5_VLForConditionalGeneration import versus a video-model processor crash without torchvision." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null }, { "label": "issue", @@ -13829,8 +13926,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -13839,55 +13936,51 @@ "issue:42490", "issue:42757", "issue:42907", + "issue:42994", "issue:43012", "issue:43064", "issue:43066", - "issue:43097", "issue:43208", "issue:43299", "issue:43352", + "issue:43475", "issue:43502", "issue:43526", "issue:43550", "issue:43582", "issue:43618", + "issue:43756", "issue:43761", "issue:43792", "issue:43824", "issue:43825", "issue:43867", - "issue:43873", + "issue:43901", "issue:43906", "issue:43976", - "issue:43986", "issue:43992", "issue:44038", - "issue:44117", "issue:44222", "issue:44230", "issue:44368", "issue:44380", "issue:44483", "issue:44484", + "issue:44485", "issue:44514", "issue:44655", "issue:44683", - "issue:44704", - "issue:45103", + "issue:45276", "issue:45290", "issue:45381", "issue:45478", "issue:45479", "issue:45753", + "issue:46077", "issue:46123", "issue:46132" ], "soft_pairs": [ - "issue:43502|issue:43992", - "issue:44704|issue:45103", - "issue:43097|issue:43873", - "issue:43986|issue:44117", - "issue:43824|issue:43986", "issue:43066|issue:43906", "issue:43208|issue:43550", "issue:43992|issue:44484", @@ -13895,8 +13988,8 @@ "issue:43012|issue:43550", "issue:43012|issue:44368", "issue:43352|issue:44380", - "issue:42907|issue:45753", "issue:42907|issue:44655", + "issue:42907|issue:45753", "issue:43792|issue:46123", "issue:43352|issue:43976", "issue:44380|issue:46132", @@ -13915,244 +14008,253 @@ "issue:44038|issue:44230", "issue:44483|issue:45290", "issue:44483|issue:45381", - "issue:43299|issue:43867" + "issue:43299|issue:43867", + "issue:43756|issue:44485", + "issue:36683|issue:43475", + "issue:43299|issue:45753", + "issue:42994|issue:44655", + "issue:43901|issue:44655", + "issue:45276|issue:46077" ], - "prepared_review_unit_hash": "ac6afa28539c0ce94d09e61ef4d6cd43257cd4e4664f6ea066a0be26ca34f540", + "prepared_review_unit_hash": "709fd6eb04bf27f0abe7c58af973776b3f587d68b367d06883265a01bfaa1fe5", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 47, - "item_count": 47, - "soft_pair_count": 33, - "serialized_chars": 36842, - "estimated_input_tokens": 9211, - "estimated_eval_tokens": 18678 + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37656, + "estimated_input_tokens": 9414, + "estimated_eval_tokens": 19084 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "Mostly a noisy cluster of unrelated Transformers issues. I only found one plausible duplicate pair: the two CLIP v5 regression reports about missing outputs. Everything else spans different model families, APIs, or bug classes and should stay separate.", - "confidence": 0.84, - "canonical_issue_reason": "No single canonical issue: the cluster is heterogeneous and does not form one duplicate set.", + "summary": "The set is very heterogeneous: most items are unrelated model-specific regressions, loading/saving issues, or API/behavior changes. None of the soft pairs look like true duplicates, so I rejected them all. If one issue must represent the cluster, the broad v5 chat-completions regression is the closest fit, but it does not explain the rest of the items.", + "confidence": 0.37, + "canonical_issue_reason": "issue:44483 is the broadest user-facing regression in the set and best matches the loose v5 API-breakage theme, though the cluster is too mixed for it to be a strong canonical duplicate.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43618 is the closest local anchor because it is paired with 43761 in the same CLIP output-regression area, but even that only covers a small subset of the cluster.", + "best_issue_reason": "issue:44483 is the most representative single issue by scope and severity, but only as a loose umbrella for the cluster\u2014not as a duplicate of the others.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43502", - "right": "issue:43992", - "accept": false, - "reason": "Different subsystems: local_files_only/network behavior vs missing checkpoint weight loading in UMT5Encoder." - }, - { - "left": "issue:44704", - "right": "issue:45103", - "accept": false, - "reason": "Different code paths: AutoProcessor kwargs passthrough vs auto-docstring annotation handling." - }, - { - "left": "issue:43097", - "right": "issue:43873", - "accept": false, - "reason": "Unrelated problems: removed config API vs offloading/quantization behavior." - }, - { - "left": "issue:43986", - "right": "issue:44117", - "accept": false, - "reason": "Different failures: missing torchvision crash in AutoProcessor vs tokenizer mapping returning None." - }, - { - "left": "issue:43824", - "right": "issue:43986", - "accept": false, - "reason": "Both are import/load issues around vision models, but they affect different symbols and failure modes." - }, { "left": "issue:43066", "right": "issue:43906", "accept": false, - "reason": "No shared concrete bug; one is tokenizer decoder typing, the other is a separate reproduction report." + "reason": "Different bugs: tokenizer decoder type vs an isolated reproduction for a separate issue." }, { "left": "issue:43208", "right": "issue:43550", "accept": false, - "reason": "Different model/backend issues: xLSTM training bugs vs Bamba torch.compile with SDPA." + "reason": "Unrelated model/training bugs; xLSTM training failures are not the same as Bamba SDPA compile failures." }, { "left": "issue:43992", "right": "issue:44484", "accept": false, - "reason": "Unrelated: missing encoder weight in UMT5 vs save_pretrained shard-size default question." + "reason": "Missing embed_tokens.weight on load is not the same as a save_pretrained max_shard_size question." }, { "left": "issue:43825", "right": "issue:44655", "accept": false, - "reason": "Different areas: pipeline error messaging vs saving Pipeline objects." + "reason": "One is a pipeline task support/error-message issue; the other is pipeline serialization. Different code paths." }, { "left": "issue:43012", "right": "issue:43550", "accept": false, - "reason": "Both mention compilation/warnings, but one is a bfloat16 warning and the other is an actual SDPA compile failure on a specific model." + "reason": "A bfloat16 warning during compilation is not the same bug as Bamba torch.compile with SDPA." }, { "left": "issue:43012", "right": "issue:44368", "accept": false, - "reason": "Different warnings with different causes: PyTorch precision warning vs tie_word_embeddings config warning." + "reason": "Both mention warnings, but they concern different configs and different behaviors." }, { "left": "issue:43352", "right": "issue:44380", "accept": false, - "reason": "Both involve attention, but one is unsupported Flash Attention for Nemotron and the other is GPT2 scaling ignored under backends." + "reason": "Different concrete failures: Flash Attention support gating vs GPT2 attention scaling being ignored under SDPA/FlashAttention." }, { "left": "issue:42907", - "right": "issue:45753", + "right": "issue:44655", "accept": false, - "reason": "Completely different bugs: saving dequantized models vs infinite-loop generation on a specific image." + "reason": "Both involve saving, but one is quantized/dequantized model saving and the other is saving Pipeline objects." }, { "left": "issue:42907", - "right": "issue:44655", + "right": "issue:45753", "accept": false, - "reason": "Save/pretrained is broad, but these are different objects and failure modes: dequantized models vs Pipeline saving." + "reason": "Completely different symptoms: model saving vs generation looping on a specific image." }, { "left": "issue:43792", "right": "issue:46123", "accept": false, - "reason": "Unrelated: Whisper model execution vs MaskGenerationPipeline partial-batch result loss." + "reason": "Whisper model loading/running issue vs MaskGenerationPipeline partial-batch result dropping." }, { "left": "issue:43352", "right": "issue:43976", "accept": false, - "reason": "Different problems: model attention backend support vs Python version compatibility." + "reason": "Flash Attention support in a model is unrelated to Python version compatibility." }, { "left": "issue:44380", "right": "issue:46132", "accept": false, - "reason": "Both mention attention, but one is GPT2 backend scaling and the other is AttentionInterface.register behavior change." + "reason": "Attention scaling being ignored is not the same as AttentionInterface.register changing function behavior." }, { "left": "issue:44380", "right": "issue:45479", "accept": false, - "reason": "Unrelated: attention scaling bug vs degenerate zero-loss classification setup." + "reason": "Attention backend bug vs sequence-classification zero-loss bug; different model behavior and code path." }, { "left": "issue:43064", "right": "issue:44380", "accept": false, - "reason": "Different code paths: FSDP2/PEFT optimizer state corruption vs GPT2 attention backend scaling." + "reason": "Distributed training optimizer-state corruption is unrelated to GPT2 attention scaling under SDPA/FlashAttention." }, { "left": "issue:44222", "right": "issue:45478", "accept": false, - "reason": "Related to MoE/FP8, but one is save_pretrained and the other is from_pretrained loading." + "reason": "Both are Qwen3.5/MoE-adjacent, but one is FP8 save_pretrained and the other is from_pretrained loading failure." }, { "left": "issue:44483", "right": "issue:44514", "accept": false, - "reason": "Different APIs: chat completions endpoint rejection vs batched chat template crash." + "reason": "Chat-completions request rejection is not the same as a Qwen2_5_VLProcessor batching crash." }, { "left": "issue:36683", "right": "issue:43582", "accept": false, - "reason": "No overlap beyond generic runtime failure; Gemma3 config missing vocab_size is unrelated to Apple Silicon allocator warmup TypeError." + "reason": "Gemma3Config missing vocab_size and Apple Silicon allocator warmup TypeError are unrelated." }, { "left": "issue:43208", "right": "issue:43526", "accept": false, - "reason": "Different model families and bugs: xLSTM training issues vs BeitImageProcessorFast label reduction." + "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are different subsystems and failures." }, { "left": "issue:43012", "right": "issue:44683", "accept": false, - "reason": "Both are compile-related, but one is a warning on bfloat16 and the other is a flex_attention failure on torch>=2.9." + "reason": "Precision warning in compilation is not the same as compiled flex_attention failing on torch>=2.9." }, { "left": "issue:42490", "right": "issue:43012", "accept": false, - "reason": "Path-ending-slash import warning is unrelated to the bfloat16 compilation warning." + "reason": "Path-with-slash import warning and bfloat16 compilation warning are unrelated." }, { "left": "issue:42757", "right": "issue:43824", "accept": false, - "reason": "Both are ImportErrors, but for different missing symbols and different dependency/version causes." + "reason": "Both are import errors, but they concern different symbols and packages." }, { "left": "issue:43618", "right": "issue:43761", - "accept": true, - "reason": "Both report the same CLIP v5 regression family: requested vision outputs are no longer populated in CLIP model outputs, so one fix to CLIP output assembly could plausibly address both." + "accept": false, + "reason": "CLIP attentions being unset is different from CLIPVisionModel hidden_states regression." }, { "left": "issue:43618", "right": "issue:44380", "accept": false, - "reason": "Different model families and bugs: CLIP output fields vs GPT2 attention scaling under SDPA/FlashAttention." + "reason": "Different attention-related bugs with different models and failure modes." }, { "left": "issue:38617", "right": "issue:43502", "accept": false, - "reason": "Different import/network issues: missing configuration_utils symbol vs API calls made despite local_files_only=True." + "reason": "ImportError from transformers.configuration_utils is unrelated to local_files_only still allowing API requests." }, { "left": "issue:44038", "right": "issue:44230", "accept": false, - "reason": "Same broad Qwen3-VL-Moe area, but one is a v5 loading bug and the other is fp8 inference support; not the same concrete change." + "reason": "Both mention Qwen3-VL-Moe/FP8, but one is a load/regression issue and the other is FP8 inference support." }, { "left": "issue:44483", "right": "issue:45290", "accept": false, - "reason": "Different request surfaces: chat completions API rejection vs apply_chat_template crash on tool-call messages." + "reason": "v1/chat/completions rejection is not the same as apply_chat_template crashing on tool-call assistant messages." }, { "left": "issue:44483", "right": "issue:45381", "accept": false, - "reason": "Different multimodal issues: chat completions request handling vs Qwen2.5-VL video vision_position_ids." + "reason": "Different layers: chat-completions request handling vs Qwen2.5-VL video position IDs." }, { "left": "issue:43299", "right": "issue:43867", "accept": false, - "reason": "Different failures around model loading: Qwen3VL MoE loading regression vs state_dict sorting error." + "reason": "Qwen3VL-Moe loading break and sorted-state_dict load error are distinct load-path bugs." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE-layer loss and GLM-5 RoPE implementation discussion are not the same concrete bug." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Missing Gemma3Config vocab_size and missing Sam3VisionEncoderOutput field are separate attribute mismatches." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Model loading regression vs infinite loop on a specific image; unrelated symptoms." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure is not the same as pipeline save_pretrained serialization." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Docs mismatch for return_all_scores is unrelated to pipeline serialization failure." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Gemma4 embedding resize behavior and bidirectional-attention mask construction are different bugs in the same family, not the same issue." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative. The only accepted soft edge, issue:43618 \u2194 issue:43761, is a plausible CLIP v5 regression pairing, and the rest of the cluster is reasonably treated as heterogeneous. No overstatement or mergeability issue detected." - }, - "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 }, @@ -14164,41 +14266,39 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ "issue:36010", + "issue:36296", "issue:36683", "issue:41669", + "issue:42490", "issue:42757", "issue:42890", "issue:42898", - "issue:42994", "issue:43010", "issue:43064", "issue:43066", - "issue:43299", "issue:43352", - "issue:43475", "issue:43525", + "issue:43575", "issue:43611", "issue:43618", - "issue:43756", "issue:43867", - "issue:43901", "issue:43931", "issue:43937", "issue:43976", "issue:43992", + "issue:44075", "issue:44077", "issue:44242", "issue:44246", "issue:44263", "issue:44355", "issue:44380", - "issue:44485", "issue:44492", "issue:44625", "issue:44655", @@ -14206,31 +14306,26 @@ "issue:44938", "issue:45071", "issue:45216", - "issue:45276", "issue:45278", "issue:45310", "issue:45357", "issue:45375", + "issue:45381", "issue:45440", + "issue:45478", "issue:45561", "issue:45593", - "issue:45753", + "issue:45850", "issue:46077", "issue:46082" ], "soft_pairs": [ - "issue:43756|issue:44485", - "issue:36683|issue:43475", - "issue:43299|issue:45753", - "issue:42994|issue:44655", - "issue:43901|issue:44655", - "issue:45276|issue:46077", "issue:36010|issue:42757", "issue:42898|issue:43066", "issue:43352|issue:43937", - "issue:43867|issue:45216", - "issue:36683|issue:44077", "issue:44077|issue:44380", + "issue:36683|issue:44077", + "issue:43867|issue:45216", "issue:43611|issue:44655", "issue:43976|issue:44938", "issue:44625|issue:46077", @@ -14250,222 +14345,241 @@ "issue:36683|issue:43064", "issue:42890|issue:45561", "issue:44938|issue:45071", - "issue:43867|issue:45357" + "issue:43867|issue:45357", + "issue:43010|issue:45850", + "issue:43992|issue:45478", + "issue:43010|issue:43976", + "issue:44938|issue:45381", + "issue:43618|issue:44077", + "issue:42490|issue:44075", + "issue:36296|issue:43575" ], - "prepared_review_unit_hash": "3527879584304bd62ed96c8b1f88a154da7c8f9c570c341701fd12192286f4a7", + "prepared_review_unit_hash": "bf12d1f4bc0b29649ffdd852c438d910f54fcec924efebcaa5529a53a399de7c", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { - "node_count": 48, - "item_count": 48, - "soft_pair_count": 32, - "serialized_chars": 36905, - "estimated_input_tokens": 9227, - "estimated_eval_tokens": 18710 + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 36528, + "estimated_input_tokens": 9132, + "estimated_eval_tokens": 18520 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "The set is highly heterogeneous: most items are distinct bug reports affecting different models, configs, or features. I found no convincing duplicate pairs among the provided soft edges.", - "confidence": 0.94, - "canonical_issue_reason": "issue:41669 is the clearest, most concrete representative issue here: a specific import-time performance regression caused by `import *` usage in models.", + "summary": "This cluster is mostly a heterogeneous mix of unrelated Transformers bug reports. The soft pairs share broad themes like import regressions, config serialization, or model-loading issues, but they do not look like the same concrete underlying bug. If one issue must represent the cluster, the Qwen3.5 save/load regression is the most concrete and actionable, but it is not a true duplicate hub for the rest.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 45216 is the most concrete regression-style report with a specific model, API, and failure mode (`save_pretrained` on Qwen3.5). It is better triage material than the broader meta/issues, though it does not unify the whole cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:41669 is the strongest global representative among the listed issues because it states a concrete root cause and an actionable performance problem.", + "best_issue_reason": "Issue 45216 is the best single representative because it is narrowly scoped, reproducible, and framed as a regression with clear user impact. The other items are either broader, older, or about different models/subsystems.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43756", - "right": "issue:44485", - "accept": false, - "reason": "Different problems: Smollm3 RoPE layer dropping vs a GLM-5 RoPE implementation discussion." - }, - { - "left": "issue:36683", - "right": "issue:43475", - "accept": false, - "reason": "Both mention missing attributes, but they are unrelated models/objects and different failing code paths." - }, - { - "left": "issue:43299", - "right": "issue:45753", - "accept": false, - "reason": "Different Qwen variants and different failures: model loading breakage vs an infinite-loop generation bug on one image." - }, - { - "left": "issue:42994", - "right": "issue:44655", - "accept": false, - "reason": "Both relate to saving, but one is quantized model serialization and the other is saving Pipeline objects." - }, - { - "left": "issue:43901", - "right": "issue:44655", - "accept": false, - "reason": "Docs/behavior mismatch for `return_all_scores` is unrelated to Pipeline `save_pretrained` failures." - }, - { - "left": "issue:45276", - "right": "issue:46077", - "accept": false, - "reason": "Different Gemma4 bugs: token embedding resize propagation vs incorrect attention-mask construction." - }, { "left": "issue:36010", "right": "issue:42757", "accept": false, - "reason": "Both are import errors, but for different symbols and different packages." + "reason": "Both are import errors, but they involve different missing symbols/modules and likely different root causes." }, { "left": "issue:42898", "right": "issue:43066", "accept": false, - "reason": "Seed-related flaky tests are unrelated to tokenizer decoder type issues." + "reason": "One is about `clean_up_tokenization_spaces` behavior, the other about tokenizer decoder type; same area, different bug." }, { "left": "issue:43352", "right": "issue:43937", "accept": false, - "reason": "FlashAttention support warning is unrelated to invalid GenerationConfig handling." + "reason": "Flash Attention support gating and invalid `GenerationConfig` are unrelated failure modes." }, { - "left": "issue:43867", - "right": "issue:45216", + "left": "issue:44077", + "right": "issue:44380", "accept": false, - "reason": "Different save/load bugs: sorted state_dict loading vs incorrect Qwen3.5 checkpoint saving." + "reason": "These touch different code paths: optimizer argument handling versus GPT-2 attention scaling under SDPA/FlashAttention." }, { "left": "issue:36683", "right": "issue:44077", "accept": false, - "reason": "Missing config attribute in Gemma3 is unrelated to patchtsmixer post_init allowance." + "reason": "Missing `vocab_size` in Gemma3Config and `patchtsmixer` post-init restrictions are unrelated config issues." }, { - "left": "issue:44077", - "right": "issue:44380", + "left": "issue:43867", + "right": "issue:45216", "accept": false, - "reason": "Patchtsmixer post_init policy and GPT2 attention scaling under SDPA/FlashAttention are different issues." + "reason": "Both concern model loading/saving, but one is a generic sorted-state_dict load failure and the other is a Qwen3.5 save_pretrained regression." }, { "left": "issue:43611", "right": "issue:44655", "accept": false, - "reason": "Base-model-prefix loading regression is unrelated to Pipeline save_pretrained serialization." + "reason": "`base_model_prefix` loading failures and pipeline serialization are different save/load paths with different symptoms." }, { "left": "issue:43976", "right": "issue:44938", "accept": false, - "reason": "Both are Python-version compatibility reports, but they refer to different versions and likely different causes." + "reason": "Python version compatibility problems, but one targets 3.10+ behavior and the other 3.14 load failure; not the same bug." }, { "left": "issue:44625", "right": "issue:46077", "accept": false, - "reason": "Qwen3.5 num_labels propagation and Gemma4 attention masks are unrelated model/config bugs." + "reason": "Qwen3.5 `num_labels` propagation and Gemma4 bidirectional attention masking are unrelated model-config bugs." }, { "left": "issue:46077", "right": "issue:46082", "accept": false, - "reason": "Different config validation and different model behavior; no shared underlying bug." + "reason": "One is an attention-mask construction bug, the other is `head_dim` validation in LlamaConfig; different issues." }, { "left": "issue:43352", "right": "issue:45593", "accept": false, - "reason": "FlashAttention support for Nemotron is unrelated to D-FINE auxiliary loss omission." + "reason": "Flash Attention support and D-FINE auxiliary losses are entirely separate model behaviors." }, { "left": "issue:36683", "right": "issue:43525", "accept": false, - "reason": "Two missing-attribute reports, but for different configs and different fields." + "reason": "Both are missing-config-attribute errors, but for different models and different fields; not the same underlying bug." }, { "left": "issue:43618", "right": "issue:45216", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to Qwen3.5 checkpoint saving." + "reason": "CLIPOutput attentions regression and Qwen3.5 save_pretrained regression are unrelated." }, { "left": "issue:44263", "right": "issue:45440", "accept": false, - "reason": "Different MoE/model-implementation issues in different architectures." + "reason": "`torch.split()` return-value handling and DeepSeekV3 MoE divergence are different implementation problems." }, { "left": "issue:36683", "right": "issue:44242", "accept": false, - "reason": "Missing `vocab_size` and missing load-balancing loss are unrelated failures." + "reason": "A missing config attribute and an MoE load-balancing loss bug are unrelated." }, { "left": "issue:43352", "right": "issue:43931", "accept": false, - "reason": "Unsupported FlashAttention and Qwen3-VL weight-shape mismatch are different problems." + "reason": "Flash Attention support error and Qwen3-VL weight-shape mismatch are different model-loading failures." }, { "left": "issue:43064", "right": "issue:44492", "accept": false, - "reason": "Distributed optimizer-state corruption is unrelated to a cache-strategy typo." + "reason": "FSDP2 optimizer state corruption and a typo in cache strategy docs/code are not the same issue." }, { "left": "issue:43010", "right": "issue:45278", "accept": false, - "reason": "Cache `update()` no_grad semantics and broad import failures are not the same bug." + "reason": "`@torch.no_grad` on cache updates and broad import failures after upgrade are too different to merge." }, { "left": "issue:43992", "right": "issue:45310", "accept": false, - "reason": "Missing `embed_tokens.weight` in UMT5 is unrelated to Qwen3.5 MoE `from_pretrained` errors." + "reason": "UMT5Encoder missing weights and Qwen3.5 MoE from_pretrained errors affect different models and code paths." }, { "left": "issue:43867", "right": "issue:45375", "accept": false, - "reason": "State-dict ordering load error is unrelated to a missing Qwen3.5 vision config field." + "reason": "Generic load-state_dict ordering failure versus missing Qwen3.5 vision config field; related serialization area, but not the same bug." }, { "left": "issue:44355", "right": "issue:44855", "accept": false, - "reason": "Compiled-file errors and a Python 3.13 JIT parsing/IndentationError are different import-time issues." + "reason": "Compiled-file import errors and a Python 3.13 `@torch.jit.script` parsing problem are different import-time failures." }, { "left": "issue:41669", "right": "issue:44246", "accept": false, - "reason": "Both concern import performance, but one has a specific root cause (`import *` in models) while the other is a generic slowdown report." + "reason": "Import performance issues are broad, but one is about removing `import *` and the other about sporadic slow imports." + }, + { + "left": "issue:36683", + "right": "issue:43064", + "accept": false, + "reason": "Gemma3 config attribute errors and FSDP2 optimizer-state corruption are unrelated." }, { "left": "issue:42890", "right": "issue:45561", "accept": false, - "reason": "Seed-related test flakiness is unrelated to xdist race conditions in captured-info handling." + "reason": "Both are test instability issues, but they involve different races and different files; not one underlying bug." }, { "left": "issue:44938", "right": "issue:45071", "accept": false, - "reason": "Python 3.14 load failure and PretrainedConfig type-checking breakage are distinct regressions." + "reason": "Python 3.14 load failure and `PretrainedConfig` type-checking regression are separate compatibility problems." }, { "left": "issue:43867", "right": "issue:45357", "accept": false, - "reason": "Different Qwen3.5 save/load regressions affecting different artifacts and key mappings." + "reason": "Both touch loading/saving, but one is a generic state_dict ordering issue and the other is a specific incorrect visual encoder key regression." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A concrete cache-update no-grad fix and a broad complaint that v5 breaks many models are not duplicates." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Different models, different missing/misconfigured weights, and different from_pretrained failures." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache mutation under no-grad and Python version support are unrelated." + }, + { + "left": "issue:44938", + "right": "issue:45381", + "accept": false, + "reason": "One is a Python 3.14 load failure; the other is a Qwen2.5-VL video position-id bug." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIP attentions missing and `patchtsmixer` post_init policy are unrelated model regressions." + }, + { + "left": "issue:42490", + "right": "issue:44075", + "accept": false, + "reason": "A trailing-slash local-model import warning and unused SGD args are unrelated." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Both mention tensor parallelism, but one is a training bug and the other is an OOM on model load; not the same fix." } ] }, @@ -14484,8 +14598,8 @@ "review_unit_count": 49, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ @@ -14494,27 +14608,28 @@ "issue:30990", "issue:31795", "issue:33453", + "issue:34689", "issue:36296", - "issue:42490", "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:43827", "issue:43867", - "issue:43976", - "issue:43992", + "issue:43874", "issue:44062", - "issue:44075", - "issue:44077", "issue:44112", "issue:44222", + "issue:44230", + "issue:44315", "issue:44351", "issue:44355", "issue:44360", @@ -14522,30 +14637,22 @@ "issue:44484", "issue:44485", "issue:44556", + "issue:44655", "issue:44749", "issue:44857", - "issue:44938", "issue:44961", "issue:44964", "issue:44977", + "issue:45071", "issue:45125", "issue:45341", "issue:45375", - "issue:45381", "issue:45440", "issue:45478", "issue:45600", - "issue:45850", "issue:46132" ], "soft_pairs": [ - "issue:43010|issue:45850", - "issue:43992|issue:45478", - "issue:43010|issue:43976", - "issue:44938|issue:45381", - "issue:43618|issue:44077", - "issue:42490|issue:44075", - "issue:36296|issue:43575", "issue:28282|issue:33453", "issue:44360|issue:45440", "issue:43867|issue:44484", @@ -14556,9 +14663,9 @@ "issue:42890|issue:45600", "issue:43502|issue:44062", "issue:43723|issue:44556", + "issue:42548|issue:44351", "issue:43576|issue:46132", "issue:42890|issue:45341", - "issue:42548|issue:44351", "issue:43010|issue:43065", "issue:44387|issue:44556", "issue:44222|issue:45125", @@ -14571,234 +14678,269 @@ "issue:45125|issue:45375", "issue:43502|issue:44222", "issue:43299|issue:44556", - "issue:43502|issue:44961" + "issue:43502|issue:44961", + "issue:44315|issue:44360", + "issue:42831|issue:43874", + "issue:44355|issue:45071", + "issue:30333|issue:36296", + "issue:43827|issue:44961", + "issue:44355|issue:44655", + "issue:44222|issue:44230", + "issue:33453|issue:34689", + "issue:42757|issue:43576", + "issue:33453|issue:36296", + "issue:43576|issue:45071" ], - "prepared_review_unit_hash": "4bdef01ea345396ec7b5d5711e23ffaa0ef9c9b35aec5f371dee3252de68f070", + "prepared_review_unit_hash": "d9fb1f837bf1757ab9f3863cbbba6de589ccf0b9870a119f56196a9b862a0c26", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, - "soft_pair_count": 33, - "serialized_chars": 36879, - "estimated_input_tokens": 9220, - "estimated_eval_tokens": 18696 + "soft_pair_count": 37, + "serialized_chars": 38147, + "estimated_input_tokens": 9537, + "estimated_eval_tokens": 19330 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is noisy and mostly composed of unrelated issues, many of them broad v5/regression reports. None of the soft pairs look like true duplicates or the same fix target.", - "confidence": 0.95, - "canonical_issue_reason": "issue:45850 is the broadest umbrella in the set: it summarizes that v5 breaks many models, while most other issues are narrower, model-specific regressions or unrelated bugs.", + "summary": "This cluster is highly heterogeneous: most items are unrelated bugs, docs, CI, or feature regressions. The only mild theme is \u201cloading/import regressions,\u201d but none of the soft pairs look like the same concrete bug or change, so I would not merge any of them. If forced to anchor the cluster, #33453 is the broadest tokenizer-loading regression.", + "confidence": 0.74, + "canonical_issue_reason": "#33453 is the broadest and most reusable anchor among the loading-related reports: it names a general tokenizer-loading regression, whereas most other candidates are narrower or point to different subsystems. It is still only a loose anchor because the cluster is not actually a duplicate set.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45850 is the best global cluster anchor because it is the only high-level tracking issue that can plausibly subsume the broader v5 breakage theme, even though it is vague.", + "best_issue_reason": "#33453 is the best single issue to represent the small loading-regression subset because it is broad, explicit, and closer in scope to the other tokenizer/model-loading complaints than the more specialized reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:43010", - "right": "issue:45850", - "accept": false, - "reason": "Different bug class: cache/layer update no_grad behavior vs a broad v5 complaint." - }, - { - "left": "issue:43992", - "right": "issue:45478", - "accept": false, - "reason": "Different models and different loading failures; not the same concrete defect." - }, - { - "left": "issue:43010", - "right": "issue:43976", - "accept": false, - "reason": "Unrelated: training cache/no_grad vs Python version compatibility." - }, - { - "left": "issue:44938", - "right": "issue:45381", - "accept": false, - "reason": "Python 3.14 import/load failure is unrelated to Qwen2.5-VL video position ids." - }, - { - "left": "issue:43618", - "right": "issue:44077", - "accept": false, - "reason": "Different components and symptoms; CLIPOutput attentions vs patchtsmixer post_init." - }, - { - "left": "issue:42490", - "right": "issue:44075", - "accept": false, - "reason": "Local model path import warning is unrelated to SGD args being ignored." - }, - { - "left": "issue:36296", - "right": "issue:43575", - "accept": false, - "reason": "Both mention tensor parallelism, but one is a generic training bug and the other is a specific OOM on a model load." - }, { "left": "issue:28282", "right": "issue:33453", "accept": false, - "reason": "PyTorch import error vs tokenizer loading regression; different code paths." + "reason": "PyTorch import error vs tokenizer loading regression; different failure mode and code path." }, { "left": "issue:44360", "right": "issue:45440", "accept": false, - "reason": "DSA indexer ReLU discussion is unrelated to DeepSeekV3 implementation divergence." + "reason": "DSA indexer/ReLU discussion is unrelated to DeepSeekV3 MoE divergence." }, { "left": "issue:43867", "right": "issue:44484", "accept": false, - "reason": "State-dict loading failure is not the same as a max_shard_size question." + "reason": "State-dict sorting load bug vs save_pretrained shard-size question; unrelated." }, { "left": "issue:43618", "right": "issue:44857", "accept": false, - "reason": "Different model failures; no shared concrete code path." + "reason": "CLIP attentions regression is unrelated to LwDetrImageLoss AMP crash." }, { "left": "issue:43618", "right": "issue:44977", "accept": false, - "reason": "CLIPOutput attention assignment and Qwen3.5 flash-attention generation are unrelated bugs." + "reason": "CLIPOutput attentions issue and Qwen flash-attention generation bug are different code paths." }, { "left": "issue:42890", "right": "issue:44964", "accept": false, - "reason": "Test seed flakiness is unrelated to loading Phi-4 multimodal with latest transformers." + "reason": "SamHQ test seeding flakiness vs Phi-4 multimodal load failure; not the same bug." }, { "left": "issue:42890", "right": "issue:44112", "accept": false, - "reason": "Both are test failures, but the underlying causes are different." + "reason": "Both are test-related, but one is seed stability and the other is a stale device override CI failure." }, { "left": "issue:42890", "right": "issue:45600", "accept": false, - "reason": "CI test flake vs auto_mappings stale config references; different fixes." + "reason": "Test seeding issue vs auto_mappings config reference breakage; unrelated." }, { "left": "issue:43502", "right": "issue:44062", "accept": false, - "reason": "local_files_only API leak is unrelated to AddedToken argument duplication." + "reason": "local_files_only network leak vs AddedToken kwarg error; different behaviors." }, { "left": "issue:43723", "right": "issue:44556", "accept": false, - "reason": "Tokenizer loading regression and checkpoint reload regression are related only at a broad level, not the same bug." + "reason": "Tokenizer loading regression and checkpoint reload regression are both loading-related, but not the same concrete failure." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Different missing exports: PreTrainedModel vs HybridCache." }, { "left": "issue:43576", "right": "issue:46132", "accept": false, - "reason": "Broken env command is unrelated to AttentionInterface.register behavior changes." + "reason": "Broken CLI env command vs changed register behavior; unrelated." }, { "left": "issue:42890", "right": "issue:45341", "accept": false, - "reason": "Different test issues and different affected code paths." - }, - { - "left": "issue:42548", - "right": "issue:44351", - "accept": false, - "reason": "Both are import errors, but for different exported symbols and likely different packaging issues." + "reason": "SamHQ seed flakiness vs generic testing_utils bug; not the same issue." }, { "left": "issue:43010", "right": "issue:43065", "accept": false, - "reason": "Cache update decoration and Sam3PixelDecoder dummy Conv2d are unrelated." + "reason": "Cache/Layers no_grad decoration request vs Sam3PixelDecoder dummy Conv2d; different subsystems." }, { "left": "issue:44387", "right": "issue:44556", "accept": false, - "reason": "CUDA memory regression under int4 quantization is not the same as checkpoint reload breakage." + "reason": "Int4 CUDA memory regression vs checkpoint reload after upgrade; different bugs." }, { "left": "issue:44222", "right": "issue:45125", "accept": false, - "reason": "Both involve Qwen3.5 MoE, but one is FP8 save_pretrained and the other is tensor-parallel plan metadata." + "reason": "FP8 save_pretrained bug vs missing tensor-parallel plan; same general area, but not one fixable PR." }, { "left": "issue:43502", "right": "issue:44485", "accept": false, - "reason": "local_files_only request leakage is unrelated to GLM-5 RoPE implementation." + "reason": "local_files_only network leak vs RoPE implementation discussion; unrelated." }, { "left": "issue:44355", "right": "issue:44749", "accept": false, - "reason": "Compiled Python file errors are unrelated to the Chinese-language v4.57-to-v5.3 slowdown report." + "reason": "Compiled Python file errors vs performance slowdown after upgrade; not the same failure." }, { "left": "issue:45125", "right": "issue:45478", "accept": false, - "reason": "Same family of models, but different failures: missing tp plan vs from_pretrained error." + "reason": "Missing _tp_plan and from_pretrained error on Qwen3.5 MoE may be related, but the reports describe different concrete breakages." }, { "left": "issue:43010", "right": "issue:43232", "accept": false, - "reason": "Cache update no_grad concern is unrelated to generation kwargs syncing." + "reason": "No-grad decoration request vs generation kwargs update bug; different code paths." }, { "left": "issue:30990", "right": "issue:31795", "accept": false, - "reason": "Sentence Transformers loading hang and documentation confusion about forward args are unrelated." + "reason": "Sentence Transformers loading hang vs documentation confusion for model.forward." }, { "left": "issue:30333", "right": "issue:33453", "accept": false, - "reason": "MLflow reporting on failed jobs is unrelated to tokenizer loading regression." + "reason": "MLflow reporting on failed jobs is unrelated to tokenizer loading regressions." }, { "left": "issue:45125", "right": "issue:45375", "accept": false, - "reason": "Different Qwen3.5 MoE/Vision config fields and different defects." + "reason": "Tensor-parallel metadata issue vs missing vision config field; related model family, different bug." }, { "left": "issue:43502", "right": "issue:44222", "accept": false, - "reason": "local_files_only network leakage and FP8 moe save_pretrained are unrelated." + "reason": "local_files_only network leak vs FP8 save_pretrained moe; different defects." }, { "left": "issue:43299", "right": "issue:44556", "accept": false, - "reason": "Qwen3VL MoE loading breakage and generic checkpoint reload breakage are not the same defect." + "reason": "Qwen3VL MoE load break and version-upgrade checkpoint reload failure are both loading issues, but not the same concrete regression." }, { "left": "issue:43502", "right": "issue:44961", "accept": false, - "reason": "No meaningful overlap; racoon is unrelated noise." + "reason": "API network leak vs a junk/non-informative issue title; no overlap." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel/model_init issue is unrelated to the DSA indexer ReLU discussion." + }, + { + "left": "issue:42831", + "right": "issue:43874", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue vs GLM46V image processor attribute error; unrelated." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file runtime errors vs PretrainedConfig type checking regression; different bugs." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLflow reporting bug vs tensor-parallel training bug; unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs still referencing pipeline() vs unrelated junk title; no shared bug." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled file execution errors vs inability to save Pipeline objects; different issues." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "FP8 save_pretrained bug vs FP8 inference support request; same broad theme, but distinct changes." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer loading regression vs Llama 3.2 vision model-loading breakage; different concrete failures." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing import from huggingface_hub vs broken transformers env CLI; unrelated." + }, + { + "left": "issue:33453", + "right": "issue:36296", + "accept": false, + "reason": "Tokenizer loading regression vs tensor parallel training bug; unrelated." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken env command vs PretrainedConfig type-checking regression; different code paths." } ] }, @@ -14822,32 +14964,31 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ - "issue:30333", "issue:31795", - "issue:33453", - "issue:34689", "issue:36296", "issue:38617", "issue:41628", "issue:41669", "issue:42371", "issue:42548", - "issue:42757", - "issue:42831", + "issue:42890", + "issue:43352", + "issue:43408", "issue:43525", "issue:43526", "issue:43572", "issue:43576", "issue:43673", "issue:43824", - "issue:43827", - "issue:43874", "issue:43901", + "issue:43976", "issue:44038", - "issue:44222", "issue:44230", + "issue:44246", "issue:44261", "issue:44273", + "issue:44295", + "issue:44297", "issue:44315", "issue:44355", "issue:44360", @@ -14856,7 +14997,7 @@ "issue:44492", "issue:44556", "issue:44623", - "issue:44655", + "issue:44704", "issue:44908", "issue:44961", "issue:45003", @@ -14864,6 +15005,7 @@ "issue:45092", "issue:45125", "issue:45230", + "issue:45278", "issue:45362", "issue:45468", "issue:45542", @@ -14872,17 +15014,6 @@ "issue:46095" ], "soft_pairs": [ - "issue:44315|issue:44360", - "issue:42831|issue:43874", - "issue:44355|issue:45071", - "issue:30333|issue:36296", - "issue:43827|issue:44961", - "issue:44355|issue:44655", - "issue:44222|issue:44230", - "issue:33453|issue:34689", - "issue:42757|issue:43576", - "issue:33453|issue:36296", - "issue:43576|issue:45071", "issue:43576|issue:43824", "issue:31795|issue:38617", "issue:41628|issue:42548", @@ -14900,8 +15031,8 @@ "issue:44273|issue:45542", "issue:44273|issue:45003", "issue:44038|issue:45125", - "issue:44393|issue:45753", "issue:44393|issue:45071", + "issue:44393|issue:45753", "issue:43576|issue:44556", "issue:44230|issue:46095", "issue:43525|issue:45850", @@ -14909,271 +15040,285 @@ "issue:44556|issue:45092", "issue:43572|issue:45850", "issue:44492|issue:45850", - "issue:42371|issue:44492" + "issue:42371|issue:44492", + "issue:45278|issue:45850", + "issue:42890|issue:44908", + "issue:44704|issue:44908", + "issue:44246|issue:44273", + "issue:43408|issue:44315", + "issue:43976|issue:44393", + "issue:43352|issue:44393", + "issue:43576|issue:44908", + "issue:44297|issue:45230", + "issue:44908|issue:45071", + "issue:44295|issue:45230" ], - "prepared_review_unit_hash": "95453811dd7a9b881da05db371626bd8191b63cfcb38f1fe0aeb5fa59ee5ef93", + "prepared_review_unit_hash": "1836fbe716b3a5430f10e94cfbbb67b2be32983e97c5441e14974148c4ac3d98", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 38, - "serialized_chars": 38341, - "estimated_input_tokens": 9586, - "estimated_eval_tokens": 19428 + "serialized_chars": 38439, + "estimated_input_tokens": 9610, + "estimated_eval_tokens": 19476 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is a loose mix of Transformers issues, mostly around v5 breakages, imports, and model/loading regressions, but the individual reports are not the same underlying bug. No soft pair looks safe to merge as a duplicate.", - "confidence": 0.63, - "canonical_issue_reason": "issue:44556 is the best representative because it is a concrete, user-facing v5 loading regression with a clear version boundary and reproducible impact on reloading checkpoints.", + "summary": "This cluster is mostly a loose mix of Transformers v5 regression reports, especially import/loading breakages; only one soft pair looks like the same underlying compatibility bug.", + "confidence": 0.73, + "canonical_issue_reason": "issue:45278 is the clearest umbrella report for the repeated post-upgrade import/regression theme in this cluster, while still being concrete and actionable.", "canonical_pr_reason": null, - "best_issue_reason": "issue:44556 is the most actionable and central among the v5 breakage reports; it captures the class of checkpoint reload failures better than the vague or highly specific alternatives.", + "best_issue_reason": "issue:45278 is the best representative issue for deduping because it captures the broad import-breakage symptom seen across several nearby reports without being as vague as the generic v5 complaints.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:44315", - "right": "issue:44360", - "accept": false, - "reason": "Different subsystems and failures: Liger/model_init vs a DSA indexer ReLU issue." - }, - { - "left": "issue:42831", - "right": "issue:43874", - "accept": false, - "reason": "Unrelated problems: FineGrainedFP8 accuracy vs a GLM46V image-processor AttributeError." - }, - { - "left": "issue:44355", - "right": "issue:45071", - "accept": false, - "reason": "One is a compiled-Python-file runtime problem; the other is a PretrainedConfig typing regression." - }, - { - "left": "issue:30333", - "right": "issue:36296", - "accept": false, - "reason": "MLFlow reporting on failed training jobs is unrelated to tensor-parallel training behavior." - }, - { - "left": "issue:43827", - "right": "issue:44961", - "accept": false, - "reason": "Docs still referencing pipeline() is unrelated to an empty/placeholder issue." - }, - { - "left": "issue:44355", - "right": "issue:44655", - "accept": false, - "reason": "Both mention loading/saving, but one is about compiled Python files and the other about saving Pipeline objects." - }, - { - "left": "issue:44222", - "right": "issue:44230", - "accept": false, - "reason": "Both mention FP8/MoE, but one is about save_pretrained and the other about inference support; not the same concrete bug." - }, - { - "left": "issue:33453", - "right": "issue:34689", - "accept": false, - "reason": "Both are loading regressions, but they affect different code paths and model families." - }, - { - "left": "issue:42757", - "right": "issue:43576", - "accept": false, - "reason": "Different import/runtime areas: huggingface_hub symbol import vs the transformers env command in v5." - }, - { - "left": "issue:33453", - "right": "issue:36296", - "accept": false, - "reason": "Tokenizer/model loading regression is unrelated to tensor-parallel training." - }, - { - "left": "issue:43576", - "right": "issue:45071", - "accept": false, - "reason": "A broken CLI command and a type-checking regression are not the same bug." - }, { "left": "issue:43576", "right": "issue:43824", "accept": false, - "reason": "The v5 env command issue is unrelated to missing a specific model export." + "reason": "Different symptoms: broken env command vs a missing model export import error." }, { "left": "issue:31795", "right": "issue:38617", "accept": false, - "reason": "Documentation confusion for forward args is unrelated to an ImportError for layer_type_validation." + "reason": "Documentation confusion about forward args is unrelated to an ImportError for layer_type_validation." }, { "left": "issue:41628", "right": "issue:42548", "accept": false, - "reason": "Both are import errors from transformers, but they are for different symbols and do not clearly share one root cause." + "reason": "Both are import-name failures, but they name different missing symbols and likely need separate export fixes." }, { "left": "issue:44961", "right": "issue:45468", "accept": false, - "reason": "'racoon' provides no evidence of the Gemma-4 audio positional encoding bug." + "reason": "No meaningful overlap; one is a nonsense title, the other is a Gemma audio positional-encoding bug." }, { "left": "issue:38617", "right": "issue:43576", "accept": false, - "reason": "ImportError for a utility symbol is unrelated to the env command breakage." + "reason": "ImportError for a missing helper is unrelated to the v5 env command regression." }, { "left": "issue:44908", "right": "issue:44961", "accept": false, - "reason": "Scheduler kwargs handling is unrelated to the placeholder issue." + "reason": "Scheduler kwargs bug and unrelated placeholder issue are not the same bug." }, { "left": "issue:44485", "right": "issue:44961", "accept": false, - "reason": "GLM-5 RoPE implementation is unrelated to the placeholder issue." + "reason": "GLM RoPE implementation discussion is unrelated to the placeholder issue." }, { "left": "issue:43673", "right": "issue:44492", "accept": false, - "reason": "Generation cache missing in chunked_prefill is unrelated to a cache-strategies typo." + "reason": "Chunked-prefill cache regression is unrelated to a typo in cache strategy docs/text." }, { "left": "issue:41628", "right": "issue:44355", "accept": false, - "reason": "Different failure modes: missing top-level export vs compiled Python file errors." + "reason": "ImportError for AutoImageProcessor is unrelated to compiled-file execution errors." }, { "left": "issue:36296", "right": "issue:43901", "accept": false, - "reason": "Tensor-parallel training bug is unrelated to docs mentioning return_all_scores." + "reason": "Tensor parallel training bug is unrelated to TextClassificationPipeline docs behavior." }, { "left": "issue:36296", "right": "issue:43526", "accept": false, - "reason": "Tensor-parallel training and BeitImageProcessorFast reduce_labels are unrelated." + "reason": "Tensor parallel training issue is unrelated to BeitImageProcessorFast label reduction." }, { "left": "issue:45362", "right": "issue:45850", "accept": false, - "reason": "A specific Qwen3.5 chat crash is not the same as a broad complaint that v5 breaks too many models." + "reason": "Specific chat crash vs broad 'v5 breaks too many models' complaint; not the same concrete bug." }, { "left": "issue:44261", "right": "issue:44360", "accept": false, - "reason": "RMS norm precision/config issue is unrelated to the DSA indexer problem." + "reason": "MLA layernorm precision issue is unrelated to the DSA ReLU discussion." }, { "left": "issue:44623", "right": "issue:45230", "accept": false, - "reason": "A concrete missing-files save_pretrained bug is not the same as a generic bug report." + "reason": "Missing processor save files is too specific to match the generic bug report." }, { "left": "issue:44273", "right": "issue:45542", "accept": false, - "reason": "Lazy loading failure is unrelated to a TensorFlow backend detection error." + "reason": "Lazy loading failure and undefined tf backend error are different code paths." }, { "left": "issue:44273", "right": "issue:45003", "accept": false, - "reason": "Lazy loading behavior and unsafe sys.modules access are different code paths." + "reason": "Lazy loading behavior and unsafe sys.modules access are related broadly, but not the same bug." }, { "left": "issue:44038", "right": "issue:45125", "accept": false, - "reason": "Both involve Qwen3/MoE, but one is a v5 compatibility bug and the other is missing tensor-parallel plan metadata." + "reason": "Qwen3-VL/Moe bug and missing tensor-parallel plan are different issues." }, { "left": "issue:44393", - "right": "issue:45753", + "right": "issue:45071", "accept": false, - "reason": "Both are Qwen multimodal issues, but one is 2D bounding-box hallucination/error and the other is an infinite loop on a specific image." + "reason": "2D bounding-box hallucination is unrelated to PretrainedConfig type-checking breakage." }, { "left": "issue:44393", - "right": "issue:45071", + "right": "issue:45753", "accept": false, - "reason": "A multimodal output bug is unrelated to PretrainedConfig type checking." + "reason": "Both are Qwen3-VL-family reports, but one is bbox output error and the other is an infinite loop on a specific image." }, { "left": "issue:43576", "right": "issue:44556", "accept": false, - "reason": "Both are v5 regressions, but one is the env CLI and the other is checkpoint reload compatibility." + "reason": "Broken env command and checkpoint reload failure are both v5 regressions but not the same code-path." }, { "left": "issue:44230", "right": "issue:46095", "accept": false, - "reason": "FP8 inference support for Qwen models is unrelated to a DeepSeekV4 weight-conversion question." + "reason": "FP8 support request and deepseekv4 weight-conversion question are not duplicates." }, { "left": "issue:43525", "right": "issue:45850", "accept": false, - "reason": "A missing pad_token_id on Llama4Config is unrelated to the broad v5 complaint." + "reason": "Missing pad_token_id in one config is not the same as a generic 'v5 breaks many models' complaint." }, { "left": "issue:41669", "right": "issue:43673", "accept": false, - "reason": "Removing import * from models is unrelated to missing GenerationMixin cache behavior." + "reason": "Import-star performance regression is unrelated to the chunked_prefill cache bug." }, { "left": "issue:44556", "right": "issue:45092", - "accept": false, - "reason": "Both are v5 loading issues, but one is generic checkpoint reload failure and the other is a specific remote-code/meta-initialization incompatibility." + "accept": true, + "reason": "Both describe Transformers v5 compatibility regressions when loading older checkpoints, specifically around meta initialization/backward compatibility." }, { "left": "issue:43572", "right": "issue:45850", "accept": false, - "reason": "StableLmConfig pad_token_idx regression is unrelated to the generic v5 complaint." + "reason": "StableLmConfig pad_token_idx regression is a concrete config bug, not the same as a broad v5 breakage report." }, { "left": "issue:44492", "right": "issue:45850", "accept": false, - "reason": "A typo in cache strategies is unrelated to the broad v5 breakage report." + "reason": "A typo in cache strategies is not the same as a generic 'many models break' complaint." }, { "left": "issue:42371", "right": "issue:44492", "accept": false, - "reason": "TF32 settings guidance is unrelated to a cache-strategies typo." + "reason": "TF32 API settings guidance is unrelated to a cache-strategy typo." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Both are broad v5 complaints, but one is specifically import errors after upgrade while the other is too vague to treat as the same bug." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Test flakiness from missing seeds is unrelated to scheduler kwargs handling." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor cached_file kwargs forwarding is unrelated to inverse_sqrt scheduler behavior." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Import latency is a performance issue; lazy loading not working is a functional regression." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "sam3 model-type warning is unrelated to Liger Kernel application in model_init." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version compatibility issue is unrelated to Qwen3-VL bbox output problems." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2 support error and Qwen3-VL bbox hallucination are different model bugs." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "Broken env command and inverse_sqrt scheduler kwargs are unrelated regressions." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer config class mismatch is unrelated to the generic bug report." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs bug is unrelated to type-checking breakage in PretrainedConfig." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Position_ids buffer access error is unrelated to the generic bug report." } ] }, - "evaluator_result": null, - "evaluator_used": false, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: the cluster is mostly a mix of v5-era regressions, with import/loading issues standing out, and the chosen representative issue is a reasonable umbrella. The soft-edge verdicts are conservative overall, with the one accepted pair (44556/45092) supported by the titles, and the rejections generally avoid overclaiming duplicates." + }, + "evaluator_used": true, "retried": false, - "accepted_nontrivial_soft_edge": false, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -15190,31 +15335,32 @@ "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-10000-413", "nodes": [ + "issue:31515", "issue:31795", "issue:33453", "issue:34689", "issue:37428", + "issue:38617", "issue:41669", "issue:42371", - "issue:42890", + "issue:42913", + "issue:43066", "issue:43298", - "issue:43352", - "issue:43408", "issue:43502", "issue:43519", + "issue:43525", + "issue:43572", "issue:43576", "issue:43606", + "issue:43618", "issue:43704", "issue:43827", "issue:43874", - "issue:43976", "issue:44077", - "issue:44246", + "issue:44230", "issue:44261", "issue:44263", "issue:44273", - "issue:44295", - "issue:44297", "issue:44315", "issue:44351", "issue:44360", @@ -15223,15 +15369,14 @@ "issue:44492", "issue:44623", "issue:44655", - "issue:44704", "issue:44829", "issue:44861", "issue:44908", "issue:44945", - "issue:45071", + "issue:44998", "issue:45092", "issue:45230", - "issue:45278", + "issue:45290", "issue:45310", "issue:45341", "issue:45468", @@ -15240,17 +15385,6 @@ "issue:46095" ], "soft_pairs": [ - "issue:45278|issue:45850", - "issue:42890|issue:44908", - "issue:44704|issue:44908", - "issue:44246|issue:44273", - "issue:43408|issue:44315", - "issue:43976|issue:44393", - "issue:43352|issue:44393", - "issue:43576|issue:44908", - "issue:44297|issue:45230", - "issue:44908|issue:45071", - "issue:44295|issue:45230", "issue:41669|issue:44351", "issue:41669|issue:45341", "issue:44492|issue:44945", @@ -15268,8 +15402,8 @@ "issue:44273|issue:44908", "issue:43519|issue:44485", "issue:43519|issue:44908", - "issue:31795|issue:34689", "issue:41669|issue:43704", + "issue:31795|issue:34689", "issue:31795|issue:37428", "issue:44315|issue:45092", "issue:44485|issue:45468", @@ -15281,483 +15415,403 @@ "issue:43519|issue:43704", "issue:43606|issue:45468", "issue:44263|issue:45468", - "issue:44261|issue:45468" + "issue:44261|issue:45468", + "issue:42913|issue:43066", + "issue:44908|issue:44998", + "issue:44492|issue:44998", + "issue:44861|issue:45290", + "issue:44998|issue:45850", + "issue:43618|issue:44998", + "issue:44998|issue:45341", + "issue:43525|issue:44861", + "issue:43572|issue:44861", + "issue:31515|issue:38617", + "issue:44230|issue:44623" ], - "prepared_review_unit_hash": "5e7bd06bd3b6c24c69725b4d9665ccf42dee64056d31b824899ab72ec13240d7", + "prepared_review_unit_hash": "8ee384339b0b393bee6350023486558cad32cdd291ab2b6992a46875ab5c909d", "component_budget": { "node_count": 413, "item_count": 413, "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 }, "budget": { "node_count": 48, "item_count": 48, "soft_pair_count": 42, - "serialized_chars": 39795, - "estimated_input_tokens": 9949, - "estimated_eval_tokens": 20154 + "serialized_chars": 39663, + "estimated_input_tokens": 9916, + "estimated_eval_tokens": 20088 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These items do not look like true duplicates; they span unrelated regressions, docs, model-loading, scheduler, and import issues. The soft matches are only superficially similar, so none should be merged. The best umbrella issue is the concrete v5 upgrade regression report.", - "confidence": 0.81, - "canonical_issue_reason": "issue:45278 is the most concrete broad umbrella in the set: it captures multiple import/model-loading regressions after the 4.57.0 -> 5.5.0 upgrade, whereas the closest alternative (issue:45850) is too vague.", + "summary": "This cluster is mostly heterogeneous: a few items share broad themes like imports, loading, or tokenizer/model regressions, but none of the soft pairs look like the same concrete bug/change. All soft edges are rejected. Issue 41669 is the broadest and most central representative among the set.", + "confidence": 0.23, + "canonical_issue_reason": "issue:41669 is the broadest, most cross-cutting item here: a package-wide import cleanup/performance report with the most soft overlaps to other import-related issues.", "canonical_pr_reason": null, - "best_issue_reason": "issue:45278 is the best representative triage target because it is broad enough to cover the surrounding upgrade-related breakages but still specific enough to be actionable.", + "best_issue_reason": "issue:41669 is the best single anchor because it is the most general infrastructural issue and the one most plausibly related to multiple neighboring reports.", "best_pr_reason": null, "soft_edge_verdicts": [ - { - "left": "issue:45278", - "right": "issue:45850", - "accept": false, - "reason": "Both are broad breakage reports, but one is a specific version-upgrade regression and the other is a vague complaint; not the same underlying bug." - }, - { - "left": "issue:42890", - "right": "issue:44908", - "accept": false, - "reason": "Different problems: flaky integration tests due to missing seeds vs. inverse_sqrt scheduler kwargs being ignored." - }, - { - "left": "issue:44704", - "right": "issue:44908", - "accept": false, - "reason": "Different code paths: cached_file kwarg forwarding in AutoProcessor vs. scheduler configuration handling." - }, - { - "left": "issue:44246", - "right": "issue:44273", - "accept": false, - "reason": "Import performance sometimes being slow is not the same as lazy loading failing functionally." - }, - { - "left": "issue:43408", - "right": "issue:44315", - "accept": false, - "reason": "Model-type mismatch warning in SAM3 tracker is unrelated to Liger Kernel not being applied in model_init." - }, - { - "left": "issue:43976", - "right": "issue:44393", - "accept": false, - "reason": "Python version compatibility failure and Qwen3-VL bbox output errors are unrelated." - }, - { - "left": "issue:43352", - "right": "issue:44393", - "accept": false, - "reason": "Flash Attention 2.0 unsupported for Nemotron is a different model-support issue than Qwen3-VL box output errors." - }, - { - "left": "issue:43576", - "right": "issue:44908", - "accept": false, - "reason": "Broken env command in v5 is unrelated to inverse_sqrt scheduler kwargs handling." - }, - { - "left": "issue:44297", - "right": "issue:45230", - "accept": false, - "reason": "Tokenizer save metadata mismatch does not match the generic bug report." - }, - { - "left": "issue:44908", - "right": "issue:45071", - "accept": false, - "reason": "Scheduler kwargs handling is unrelated to PretrainedConfig type-checking regressions." - }, - { - "left": "issue:44295", - "right": "issue:45230", - "accept": false, - "reason": "position_ids buffer access error is not the same as the generic bug report." - }, { "left": "issue:41669", "right": "issue:44351", "accept": false, - "reason": "Import-star performance and HybridCache import failure are different import-related issues." + "reason": "Import-star cleanup/perf issue vs a specific missing HybridCache import/export error; same area, different bug." }, { "left": "issue:41669", "right": "issue:45341", "accept": false, - "reason": "Import-star performance issue is unrelated to the testing_utils bug." + "reason": "Broad import cleanup vs a small testing_utils bug; no shared failure mode." }, { "left": "issue:44492", "right": "issue:44945", "accept": false, - "reason": "A typo in cache strategies is not the same as incorrect output under pipeline parallelism." + "reason": "Typo in cache strategies vs incorrect pipeline-parallel LLM output; unrelated code paths." }, { "left": "issue:43874", "right": "issue:45850", "accept": false, - "reason": "Missing image-patch method in GLM46V is a specific API bug, not the same as a vague broad regression complaint." + "reason": "Specific GLM46V multimodal method bug vs a generic 'many models break' report; not the same concrete issue." }, { "left": "issue:44393", "right": "issue:45478", "accept": false, - "reason": "Qwen3-VL 2D box hallucination differs from the Qwen3.5-Moe from_pretrained regression." + "reason": "Different Qwen3-VL/Gemma model-specific failures with different symptoms and fixes." }, { "left": "issue:44077", "right": "issue:44861", "accept": false, - "reason": "Optional post_init policy for patchtsmixer is unrelated to tied-weight key handling crashing on list input." + "reason": "PatchTSMixer post-init restriction vs tied-weight-key attribute crash; unrelated utilities." }, { "left": "issue:43827", "right": "issue:44908", "accept": false, - "reason": "Docs still mentioning pipeline() after removals is not the same as scheduler kwargs being ignored." + "reason": "Docs referencing removed pipeline() API vs scheduler kwargs not being passed; docs issue vs runtime logic bug." }, { "left": "issue:31795", "right": "issue:33453", "accept": false, - "reason": "Documentation confusion about forward arguments is unrelated to tokenizer loading regression." + "reason": "Forward-doc confusion vs tokenizer-loading regression; both mention loading/docs, but not the same bug." }, { "left": "issue:41669", "right": "issue:44908", "accept": false, - "reason": "Import-star performance issue and scheduler kwargs bug are different subsystems and failures." + "reason": "Import-star cleanup vs scheduler configuration propagation; unrelated fixes." }, { "left": "issue:44315", "right": "issue:44829", "accept": false, - "reason": "Liger Kernel application in model_init and flash_attention_3 degenerate training are different problems." + "reason": "Liger-kernel application during model_init vs flash_attention_3 training degeneration; different mechanisms and symptoms." }, { "left": "issue:43576", "right": "issue:44273", "accept": false, - "reason": "Broken env command and lazy loading failure are not the same underlying issue." + "reason": "Broken env CLI vs lazy-loading malfunction; different features entirely." }, { "left": "issue:44360", "right": "issue:45468", "accept": false, - "reason": "DSA indexer lacking ReLU is unrelated to Gemma4 audio positional encoding." + "reason": "DSA indexer ReLU discussion vs Gemma-4 audio positional encoding bug; unrelated internals." }, { "left": "issue:44655", "right": "issue:44861", "accept": false, - "reason": "Pipeline save_pretrained failure and tied-weight-key AttributeError are distinct save/model-internal bugs." + "reason": "save_pretrained() pipeline-saving issue vs tied-weight-key crash; save/export path vs weight-tie handling." }, { "left": "issue:45230", "right": "issue:45310", "accept": false, - "reason": "Generic bug report and Qwen3.5-Moe from_pretrained error are not the same issue." + "reason": "Generic bug report placeholder vs a specific Qwen3.5 MoE from_pretrained regression; not duplicates." }, { "left": "issue:44273", "right": "issue:44908", "accept": false, - "reason": "Lazy loading not working and scheduler kwargs ignored affect different code paths." + "reason": "Lazy loading regression vs inverse_sqrt scheduler kwargs bug; unrelated subsystems." }, { "left": "issue:43519", "right": "issue:44485", "accept": false, - "reason": "Incorrect timestamp calculation and GLM-5 RoPE implementation are unrelated model-processing bugs." + "reason": "Qwen3VL timestamp calculation bug vs GLM-5 RoPE implementation discussion; different model families and code paths." }, { "left": "issue:43519", "right": "issue:44908", "accept": false, - "reason": "Timestamp processing error has nothing to do with scheduler configuration." + "reason": "Timestamp math bug vs scheduler kwargs handling; no common underlying change." }, { - "left": "issue:31795", - "right": "issue:34689", + "left": "issue:41669", + "right": "issue:43704", "accept": false, - "reason": "Forward-argument documentation confusion is unrelated to Llama 3.2 Vision model loading breakage." + "reason": "Import cleanup/perf issue vs VRAM leak in Qwen3ForCausalLM threads; unrelated." }, { - "left": "issue:41669", - "right": "issue:43704", + "left": "issue:31795", + "right": "issue:34689", "accept": false, - "reason": "Import-star performance issue is not the same as VRAM leakage across dataloader threads." + "reason": "Documentation confusion vs Llama 3.2 model loading regression; different failure modes." }, { "left": "issue:31795", "right": "issue:37428", "accept": false, - "reason": "Documentation confusion about model.forward is unrelated to the missing _flash_supports_window_size import error." + "reason": "Forward-doc confusion vs missing flash-attention utility import; unrelated." }, { "left": "issue:44315", "right": "issue:45092", "accept": false, - "reason": "Liger Kernel not applied via model_init and old InternVL2 meta-init incompatibility are different regressions." + "reason": "Liger-kernel/model_init issue vs remote-code meta-init incompatibility for InternVL2; both init-related but different concrete problems." }, { "left": "issue:44485", "right": "issue:45468", "accept": false, - "reason": "GLM-5 RoPE implementation discussion is unrelated to Gemma4 audio positional encoding." + "reason": "Both are RoPE/positional-encoding themed, but they target different models and implementations." }, { "left": "issue:43704", "right": "issue:44485", "accept": false, - "reason": "VRAM leak in multi-threaded dataloading is unrelated to RoPE implementation details." + "reason": "VRAM leak under threaded dataloading vs RoPE implementation discussion; unrelated." }, { "left": "issue:43298", "right": "issue:46095", "accept": false, - "reason": "Continuous batching support for audio models is unrelated to asking for a weight conversion script." + "reason": "Continuous batching audio-model support vs asking for a weight-conversion script; unrelated." }, { "left": "issue:42371", "right": "issue:43704", "accept": false, - "reason": "TF32 API settings discussion is unrelated to VRAM leaks in multi-threaded inference." + "reason": "TF32 settings guidance vs VRAM leak; unrelated topics." }, { "left": "issue:43502", "right": "issue:43519", "accept": false, - "reason": "Unexpected API requests despite local_files_only and timestamp calculation are different failures." + "reason": "local_files_only network requests vs timestamp calculation bug; different code paths." }, { "left": "issue:43519", "right": "issue:44623", "accept": false, - "reason": "Timestamp calculation bug does not match processor.save_pretrained missing files." + "reason": "Timestamp calculation bug vs processor.save_pretrained missing files; unrelated processor behaviors." }, { "left": "issue:43519", "right": "issue:43704", "accept": false, - "reason": "Timestamp calculation and multi-thread VRAM leak are unrelated." + "reason": "Timestamp calculation bug vs threaded VRAM leak; no shared root cause." }, { "left": "issue:43606", "right": "issue:45468", "accept": false, - "reason": "CPU offload device mismatch for bark-small is unrelated to Gemma4 audio positional encoding." + "reason": "CPU offload device mismatch for bark-small vs Gemma-4 audio positional encoding; only loosely audio-related." }, { "left": "issue:44263", "right": "issue:45468", "accept": false, - "reason": "torch.split return semantics in GlmMoeDsaIndexer are unrelated to Gemma4 audio positional encoding." + "reason": "GlmMoeDsaIndexer split-values issue vs Gemma-4 audio positional encoding bug; unrelated." }, { "left": "issue:44261", "right": "issue:45468", "accept": false, - "reason": "MLA q_a_layernorm epsilon precision issue is unrelated to Gemma4 audio positional encoding." - } - ] - }, - "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": 49, - "status": "reviewed", - "reason": null, - "source": "cache", - "cache_hit": true, - "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-10000-413", - "nodes": [ - "issue:30333", - "issue:31515", - "issue:31795", - "issue:33453", - "issue:36296", - "issue:38617", - "issue:42913", - "issue:43066", - "issue:43298", - "issue:43525", - "issue:43572", - "issue:43618", - "issue:44230", - "issue:44492", - "issue:44623", - "issue:44861", - "issue:44908", - "issue:44998", - "issue:45290", - "issue:45310", - "issue:45341", - "issue:45850" - ], - "soft_pairs": [ - "issue:42913|issue:43066", - "issue:44908|issue:44998", - "issue:44492|issue:44998", - "issue:44861|issue:45290", - "issue:44998|issue:45850", - "issue:43618|issue:44998", - "issue:44998|issue:45341", - "issue:43525|issue:44861", - "issue:43572|issue:44861", - "issue:31515|issue:38617", - "issue:44230|issue:44623", - "issue:30333|issue:31515", - "issue:31515|issue:36296", - "issue:31515|issue:31795", - "issue:44230|issue:45310", - "issue:31515|issue:33453", - "issue:43298|issue:44230" - ], - "prepared_review_unit_hash": "9be4928e839f5bef3d2f32af091f898ff0f92105c41dc149daed8fa5dd3fc1f6", - "component_budget": { - "node_count": 413, - "item_count": 413, - "soft_pair_count": 1547, - "serialized_chars": 797926, - "estimated_input_tokens": 199482, - "estimated_eval_tokens": 399220 - }, - "budget": { - "node_count": 22, - "item_count": 22, - "soft_pair_count": 17, - "serialized_chars": 17358, - "estimated_input_tokens": 4340, - "estimated_eval_tokens": 8936 - }, - "overflow_policy": "truncate_then_skip", - "trimmed": false, - "aggressively_trimmed": false, - "split": true, - "analyst_result": { - "summary": "This cluster is a grab bag of unrelated issues across tokenizer behavior, config attributes, loading regressions, docs, training bugs, and one irrelevant/spam-like item. I do not see any soft pair that should be merged as the same underlying bug, so there is no clear canonical issue for the cluster.", - "confidence": 0.97, - "canonical_issue_reason": null, - "canonical_pr_reason": null, - "best_issue_reason": null, - "best_pr_reason": null, - "soft_edge_verdicts": [ + "reason": "MLA q_a_layernorm epsilon precision mismatch vs Gemma-4 audio positional encoding; different components." + }, { "left": "issue:42913", "right": "issue:43066", "accept": false, - "reason": "Both mention tokenizer/v5 behavior, but one is a broad version-behavior complaint and the other is a specific decoder-type issue; not enough evidence they are the same bug." + "reason": "General tokenizer behavior drift from v4 to v5 vs wrong tokenizer decoder type in v5; related theme, but not the same concrete bug." }, { "left": "issue:44908", "right": "issue:44998", "accept": false, - "reason": "Inverse-square scheduler kwargs bug is unrelated to the unrelated 'Unemployment' issue." + "reason": "Scheduler kwargs bug vs an unrelated placeholder issue title; no connection." }, { "left": "issue:44492", "right": "issue:44998", "accept": false, - "reason": "A cache-strategy typo is unrelated to the unrelated 'Unemployment' issue." + "reason": "Cache-strategy typo vs unrelated placeholder issue title." }, { "left": "issue:44861", "right": "issue:45290", "accept": false, - "reason": "Different code paths: tied-weight key handling versus chat template tool-call crashing." + "reason": "Tied-weight-key AttributeError vs chat template crash on tool-call assistant messages; different subsystems." }, { "left": "issue:44998", "right": "issue:45850", "accept": false, - "reason": "'Unemployment' is not the same bug as a generic post-5.0 model breakage report." + "reason": "Unrelated placeholder issue vs generic 'too many models break' complaint." }, { "left": "issue:43618", "right": "issue:44998", "accept": false, - "reason": "CLIPOutput attentions regression is unrelated to the unrelated 'Unemployment' issue." + "reason": "CLIPOutput attentions regression vs unrelated placeholder issue." }, { "left": "issue:44998", "right": "issue:45341", "accept": false, - "reason": "Testing-utils bug is unrelated to the unrelated 'Unemployment' issue." + "reason": "Unrelated placeholder issue vs small testing_utils bug." }, { "left": "issue:43525", "right": "issue:44861", "accept": false, - "reason": "Llama4Config pad_token_id missing and tied-weight-key AttributeError are different failures in different components." + "reason": "Llama4Config pad_token_id missing vs tied-weight-key crash; different model/config bugs." }, { "left": "issue:43572", "right": "issue:44861", "accept": false, - "reason": "StableLmConfig pad_token_idx missing is unrelated to tied-weight-key handling." + "reason": "StableLmConfig pad_token_idx missing vs tied-weight-key crash; unrelated." }, { "left": "issue:31515", "right": "issue:38617", "accept": false, - "reason": "Checkpoint loading slowness and an ImportError for layer_type_validation are different problems." + "reason": "Slow from_pretrained checkpoint loading vs missing layer_type_validation import; unrelated import/load issues." }, { "left": "issue:44230", "right": "issue:44623", "accept": false, - "reason": "FP8 support for Qwen3-VL/Qwen3.5 MoE is unrelated to processor.save_pretrained missing files." - }, + "reason": "FP8 Qwen3-VL support request vs processor.save_pretrained missing files; feature request vs 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": 49, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:36296", + "issue:43298", + "issue:44230", + "issue:45310" + ], + "soft_pairs": [ + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230" + ], + "prepared_review_unit_hash": "25f2fb61eb2bd20385176d0c5a98b45c11dc4c910deeb14ab4d257c7d1037ebc", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 6, + "serialized_chars": 6307, + "estimated_input_tokens": 1577, + "estimated_eval_tokens": 3410 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: training/MLFlow reporting, checkpoint loading performance, documentation, tokenizer regression, tensor-parallel training, audio continuous batching, fp8 support, and a Qwen3.5 MoE load error. None of the soft pairs look like true duplicates. The most suitable issue to anchor the cluster is the concrete version-specific bug report in 45310.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45310 is the clearest concrete bug report in the set: it names a specific failing path, model, and version regression. The other issues are either feature requests, docs, performance, or unrelated subsystems.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the most actionable and specific issue, with a direct bug/regression framing and a narrow reproduction surface. It is the strongest candidate to represent the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ { "left": "issue:30333", "right": "issue:31515", "accept": false, - "reason": "MLFlow reporting on failed training jobs is unrelated to slow from_pretrained checkpoint loading." + "reason": "Both are about training/loading, but one is an MLFlow status/reporting failure after job failure and the other is slow checkpoint loading. Different code paths and different symptoms." }, { "left": "issue:31515", "right": "issue:36296", "accept": false, - "reason": "Checkpoint loading performance and tensor-parallel training behavior are unrelated." + "reason": "Checkpoint-loading slowness is a performance problem; tensor parallel training bug is a separate training-parallelism issue. Not the same underlying defect." }, { "left": "issue:31515", "right": "issue:31795", "accept": false, - "reason": "Loading speed regression is not the same as confusing model.forward documentation." + "reason": "One is slow checkpoint loading; the other is confusing forward-docs for input_ids/past_key_values. Same broad area only at a very high level." }, { "left": "issue:44230", "right": "issue:45310", "accept": false, - "reason": "Both mention Qwen3.5 MoE, but one is fp8 inference support and the other is a from_pretrained loading error; different concrete issues." + "reason": "Both mention Qwen3.5 MoE, but one is fp8 inference support and the other is a from_pretrained regression on transformers>=5.4.0. Different change requests and different failure modes." }, { "left": "issue:31515", "right": "issue:33453", "accept": false, - "reason": "A slow checkpoint load report is not the same as a tokenizer-loading regression." + "reason": "Both involve loading, but one is a performance complaint and the other is a tokenizer-loading regression. Not the same concrete bug." }, { "left": "issue:43298", "right": "issue:44230", "accept": false, - "reason": "Continuous batching lacking audio-model support is unrelated to FP8 Qwen3 VL/MoE support." + "reason": "Continuous batching not supporting audio models is a serving capability gap; fp8 Qwen3/VL/MoE support is an inference-format/model support request. Different underlying work." } ] }, @@ -15776,8 +15830,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-13244-3", "nodes": [ @@ -15789,12 +15843,12 @@ "issue:13244|issue:16618", "issue:15980|issue:16618" ], - "prepared_review_unit_hash": "b3abd698a96446e8ffc48d28a060dfbe2e026d0180dba16d32664dedfc89b5a8", + "prepared_review_unit_hash": "1118fc75400c30b705a6ee05dc49d60d79e06a9989d9cbdcc3b729f0c30480fe", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2479, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -15802,7 +15856,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 2, - "serialized_chars": 2479, + "serialized_chars": 2478, "estimated_input_tokens": 620, "estimated_eval_tokens": 1496 }, @@ -15811,24 +15865,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "No strong duplicate cluster here. The Tapas tokenization issue is clearly unrelated. The two model-loading issues are both about loading/downloading models, but one is specifically an unauthenticated private-model error message and the other is a generic failure to load bert-base-uncased, so they do not look like the same underlying bug.", - "confidence": 0.82, - "canonical_issue_reason": "Issue #16618 is the broadest and most representative of the model-loading failure reports in this set, with more discussion activity than #15980.", + "summary": "Mostly a loose model-loading/support cluster, with one clearly unrelated tokenization issue. The two download/load complaints are superficially similar but not enough to treat as duplicates from the available evidence.", + "confidence": 0.63, + "canonical_issue_reason": "Issue #16618 is the broadest and most representative model-loading complaint in the set; it best fits the likely shared theme of failure to load a pretrained model.", "canonical_pr_reason": null, - "best_issue_reason": "#16618 is the best issue candidate because it is the most general user-facing failure among the issues shown and has the strongest discussion signal.", + "best_issue_reason": "#16618 is the best global issue candidate because it is the clearest central user-facing failure in this cluster, whereas #15980 is specifically about auth/error messaging and #13244 is unrelated tokenization behavior.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:13244", "right": "issue:16618", "accept": false, - "reason": "Different problems: Tapas tokenization mismatch versus a model-loading failure for bert-base-uncased. No indication they share the same code path." + "reason": "Tapas tokenization mismatch is a different problem from failing to load bert-base-uncased; no evidence of the same underlying bug or code path." }, { "left": "issue:15980", "right": "issue:16618", "accept": false, - "reason": "Related broadly to model access, but one is a private-model auth/error-message problem and the other is a generic inability to load bert-base-uncased. Not the same underlying bug." + "reason": "Both mention model download/load problems, but #15980 is specifically an unauthenticated private-model error message issue, not the same concrete failure as #16618." } ] }, @@ -15847,8 +15901,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-26350-2", "nodes": [ @@ -15858,7 +15912,7 @@ "soft_pairs": [ "issue:26350|issue:33260" ], - "prepared_review_unit_hash": "1ba3bf9742b68babdd99d85e1849a5be9160a17daeb9bcdefb7ef33275e517f0", + "prepared_review_unit_hash": "429e93d0948115d40cf999ec3818d374c39600269ca89cf434fc23b883b3fe3c", "component_budget": { "node_count": 2, "item_count": 2, @@ -15880,18 +15934,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate feature-request issues in the same broad area (community contribution / architecture support), but they target different integrations: Flash Attention 2 vs GGUF. They should not be clustered as duplicates.", - "confidence": 0.96, - "canonical_issue_reason": "Neither issue is a duplicate of the other; if one must be retained as a representative, issue #26350 is slightly broader/older and has more discussion activity.", + "summary": "The two issues are related only at a high level (architecture support requests), but they target different features: Flash Attention 2 vs GGUF. They should not be treated as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical duplicate here: the issues ask for different capabilities and would require different implementation work.", "canonical_pr_reason": null, - "best_issue_reason": "Issue #26350 is the better anchor only because it is older and has somewhat higher discussion activity, but it does not subsume #33260.", + "best_issue_reason": "26350 is the more established thread by age and discussion volume, but it is not a substitute for 33260 because the feature request is different.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:26350", "right": "issue:33260", "accept": false, - "reason": "Both are generic community-contribution feature requests, but they ask for different functionality in different subsystems (Flash Attention 2 support vs GGUF support). They are related in theme, not the same underlying change or bug." + "reason": "Different underlying requests: Flash Attention 2 support for more architectures vs GGUF support for more architectures. Similar subsystem, but not the same change." } ] }, @@ -15910,8 +15964,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-27088-5", "nodes": [ @@ -15933,12 +15987,12 @@ "issue:43834|issue:44365", "issue:44365|issue:44370" ], - "prepared_review_unit_hash": "adc4a8833c5442d08e97ea0fe7768421e3cca708def19f12f11a6b3f2d5363a4", + "prepared_review_unit_hash": "87e9a2323d8383181f033d99506b96e0f7a28838c7b3fe4ac8eafea24559c3b3", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 10, - "serialized_chars": 6443, + "serialized_chars": 6442, "estimated_input_tokens": 1611, "estimated_eval_tokens": 3478 }, @@ -15946,7 +16000,7 @@ "node_count": 5, "item_count": 5, "soft_pair_count": 10, - "serialized_chars": 6443, + "serialized_chars": 6442, "estimated_input_tokens": 1611, "estimated_eval_tokens": 3478 }, @@ -15955,30 +16009,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is a mix of distinct translation requests (Turkish, Spanish) and three near-identical generic i18n template issues. Only the generic template trio looks like true duplicates; the language-specific issues are separate requests.", + "summary": "This cluster is not a duplicate set: the visible language-specific issues are for different translations, and the near-identical placeholder issues lack enough evidence to safely merge as the same request.", "confidence": 0.93, - "canonical_issue_reason": "Issue 43834 is the best canonical representative of the duplicate subset because it is the earliest of the three near-identical generic template issues. The Turkish and Spanish issues are different language-specific translation requests, not duplicates of that template trio.", + "canonical_issue_reason": "No safe canonical issue: the concrete requests are for different languages (Turkish vs Spanish), and the generic placeholder issues do not provide enough body/target evidence to prove they are the same underlying task.", "canonical_pr_reason": null, - "best_issue_reason": "43834 is the strongest single issue to keep as the representative duplicate because it is earliest and matches the other two generic placeholder issues almost exactly.", + "best_issue_reason": "No single issue is suitable as a global best duplicate target because these look like separate translation requests rather than one shared bug/change.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:27088", "right": "issue:44370", "accept": false, - "reason": "Turkish-specific translation request vs a generic placeholder template; not the same underlying issue." + "reason": "Different translation request; only a generic placeholder title is shared, with no evidence they target the same language or underlying task." }, { "left": "issue:27088", "right": "issue:44365", "accept": false, - "reason": "Turkish-specific request and generic template issue are not the same translation task." + "reason": "Turkish translation request vs generic placeholder issue; not enough evidence of the same concrete request." }, { "left": "issue:27088", "right": "issue:43834", "accept": false, - "reason": "Specific Turkish translation request does not match the generic template issue." + "reason": "Turkish request and generic placeholder do not clearly describe the same translation work." }, { "left": "issue:27088", @@ -15990,47 +16044,44 @@ "left": "issue:28936", "right": "issue:43834", "accept": false, - "reason": "Spanish-specific translation request does not duplicate the generic template issue." + "reason": "Spanish translation request is distinct from the generic placeholder; no clear same underlying task." }, { "left": "issue:28936", "right": "issue:44365", "accept": false, - "reason": "Spanish-specific request and generic template issue are distinct tasks." + "reason": "Different translation request; the placeholder title is not enough to merge them." }, { "left": "issue:28936", "right": "issue:44370", "accept": false, - "reason": "Spanish-specific request and generic template issue are distinct tasks." + "reason": "Spanish request and generic placeholder are not shown to be the same issue." }, { "left": "issue:43834", "right": "issue:44370", - "accept": true, - "reason": "Near-identical generic i18n template title/body; same underlying duplicate issue." + "accept": false, + "reason": "Near-identical placeholder titles, but no body/target evidence proving the same language or same change." }, { "left": "issue:43834", "right": "issue:44365", - "accept": true, - "reason": "Near-identical generic i18n template issue; appears to be the same request duplicated." + "accept": false, + "reason": "Generic placeholder issues may represent separate language requests; identical titles alone are insufficient to deduplicate." }, { "left": "issue:44365", "right": "issue:44370", - "accept": true, - "reason": "Both are the same generic placeholder translation issue and look interchangeable." + "accept": false, + "reason": "The issues are indistinguishable by title, but there is no concrete evidence they are the same underlying translation request." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "Grounded and conservative overall. The summary correctly separates the Turkish and Spanish requests from the nearly identical generic template trio, and the accepted duplicate pairs are supported by the identical titles and very high similarity among 43834/44365/44370." - }, - "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 }, @@ -16042,8 +16093,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-36817-16", "nodes": [ @@ -16102,12 +16153,12 @@ "issue:43837|issue:45177", "issue:45322|issue:46133" ], - "prepared_review_unit_hash": "2f362bd06cdc7a5262285546eb781e88c2021abfe32d3ea45b3ca55158053a4c", + "prepared_review_unit_hash": "0cbd7832eec0b27ee8685d9fe7a6bb9dc9d27713b7d6db78400976c423d9475a", "component_budget": { "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21657, + "serialized_chars": 21659, "estimated_input_tokens": 5415, "estimated_eval_tokens": 11086 }, @@ -16115,7 +16166,7 @@ "node_count": 16, "item_count": 16, "soft_pair_count": 36, - "serialized_chars": 21657, + "serialized_chars": 21659, "estimated_input_tokens": 5415, "estimated_eval_tokens": 11086 }, @@ -16124,228 +16175,228 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are all distinct model-request issues across different model families and modalities; there is no strong duplicate cluster here.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "This cluster is a collection of distinct model-request issues, mostly asking to add support for different architectures or variants. They are similar in wording but not the same underlying request, so no soft-edge duplicates are accepted.", + "confidence": 0.44, + "canonical_issue_reason": "Issue 45322 is a clear, well-scoped model request with a detailed title and body, making it a reasonable representative for this broadly similar cluster.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43671 is a clear, well-scoped representative model-request issue, but it is not a duplicate of the others.", + "best_issue_reason": "Issue 45322 is the strongest single anchor here because it is specific, self-contained, and clearly formatted as a model-add request.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:36817", "right": "issue:41211", "accept": false, - "reason": "Different model families: EuroBert config vs DEIMv2." + "reason": "Different models: EuroBert vs DEIMv2. Similar request pattern, not the same change." }, { "left": "issue:36817", "right": "issue:43671", "accept": false, - "reason": "EuroBert config request is unrelated to Qwen3-TTS support." + "reason": "Different models/tasks: EuroBert vs Qwen3-TTS. Not a duplicate request." }, { "left": "issue:36817", "right": "issue:36879", "accept": false, - "reason": "EuroBert config and RF-DETR are different model additions." + "reason": "Different models: EuroBert vs RF-DETR. Same broad theme only." }, { "left": "issue:36879", "right": "issue:41211", "accept": false, - "reason": "RF-DETR and DEIMv2 are distinct detector/model requests." + "reason": "Different models: RF-DETR vs DEIMv2." }, { "left": "issue:36879", "right": "issue:43909", "accept": false, - "reason": "RF-DETR vs LFM2.5 Audio: different model and modality." + "reason": "Different models: RF-DETR vs LFM2.5 Audio 1.5B." }, { "left": "issue:36879", "right": "issue:44121", "accept": false, - "reason": "Detector request vs sparse transformer request; different change." + "reason": "Different models: RF-DETR vs OpenAI weight-sparse transformer." }, { "left": "issue:36879", "right": "issue:43837", "accept": false, - "reason": "RF-DETR and Qwen3-ASR are unrelated model additions." + "reason": "Different models/tasks: RF-DETR vs Qwen3-ASR." }, { "left": "issue:36879", "right": "issue:43671", "accept": false, - "reason": "RF-DETR and Qwen3-TTS are different models and tasks." + "reason": "Different models/tasks: RF-DETR vs Qwen3-TTS." }, { "left": "issue:37015", "right": "issue:43837", "accept": false, - "reason": "NeoBERT and Qwen3-ASR target different code paths." + "reason": "Different models/tasks: NeoBERT vs Qwen3-ASR." }, { "left": "issue:37015", "right": "issue:44121", "accept": false, - "reason": "NeoBERT vs weight-sparse transformer; not the same request." + "reason": "Different models: NeoBERT vs circuit-sparsity transformer." }, { "left": "issue:37015", "right": "issue:43909", "accept": false, - "reason": "NeoBERT and LFM2.5 Audio are unrelated model requests." + "reason": "Different models/tasks: NeoBERT vs LFM2.5 Audio 1.5B." }, { "left": "issue:36879", "right": "issue:37015", "accept": false, - "reason": "RF-DETR and NeoBERT are different model additions." + "reason": "Different models: RF-DETR vs NeoBERT." }, { "left": "issue:37015", "right": "issue:42503", "accept": false, - "reason": "NeoBERT and ModernVBERT are separate model families." + "reason": "Different models: NeoBERT vs ModernVBERT." }, { "left": "issue:41115", "right": "issue:43837", "accept": false, - "reason": "MiniCPM3 architecture and Qwen3-ASR are different features." + "reason": "Different models/tasks: MiniCPM3 architecture vs Qwen3-ASR." }, { "left": "issue:41115", "right": "issue:44121", "accept": false, - "reason": "MiniCPM3 and sparse transformer requests are unrelated." + "reason": "Different models: MiniCPM3 vs circuit-sparsity transformer." }, { "left": "issue:41115", "right": "issue:42503", "accept": false, - "reason": "MiniCPM3 and ModernVBERT are distinct model requests." + "reason": "Different models: MiniCPM3 vs ModernVBERT." }, { "left": "issue:41115", "right": "issue:45027", "accept": false, - "reason": "MiniCPM3 vs Voxtral TTS support: different models/tasks." + "reason": "Different models/tasks: MiniCPM3 vs Voxtral TTS." }, { "left": "issue:41211", "right": "issue:43671", "accept": false, - "reason": "DEIMv2 and Qwen3-TTS are different model additions." + "reason": "Different models/tasks: DEIMv2 vs Qwen3-TTS." }, { "left": "issue:42503", "right": "issue:43837", "accept": false, - "reason": "ModernVBERT and Qwen3-ASR are unrelated." + "reason": "Different models/tasks: ModernVBERT vs Qwen3-ASR." }, { "left": "issue:42503", "right": "issue:43671", "accept": false, - "reason": "ModernVBERT and Qwen3-TTS are separate requests." + "reason": "Different models/tasks: ModernVBERT vs Qwen3-TTS." }, { "left": "issue:42503", "right": "issue:42738", "accept": false, - "reason": "ModernVBERT and BERT-like models with RoPE are not the same change." + "reason": "Different requests: ModernVBERT models vs generic BERT-like models with RoPE." }, { "left": "issue:42738", "right": "issue:45027", "accept": false, - "reason": "BERT-like RoPE support and Voxtral TTS are unrelated." + "reason": "Different requests: BERT-like RoPE models vs Voxtral TTS." }, { "left": "issue:43671", "right": "issue:43837", "accept": false, - "reason": "Qwen3-TTS and Qwen3-ASR share a family but request different modality support." + "reason": "Related family, but different concrete support: Qwen3-TTS vs Qwen3-ASR." }, { "left": "issue:43671", "right": "issue:45027", "accept": false, - "reason": "Different speech models: Qwen3-TTS vs Voxtral-4B-TTS-2603." + "reason": "Different models/tasks: Qwen3-TTS vs Voxtral TTS." }, { "left": "issue:43671", "right": "issue:45177", "accept": false, - "reason": "Qwen3-TTS and DeepSeek-OCR2 are unrelated models." + "reason": "Different models: Qwen3-TTS vs DeepSeek-OCR2." }, { "left": "issue:43671", "right": "issue:44121", "accept": false, - "reason": "TTS support and sparse transformer support are different code changes." + "reason": "Different models/tasks: Qwen3-TTS vs circuit-sparsity transformer." }, { "left": "issue:43671", "right": "issue:43909", "accept": false, - "reason": "Qwen3-TTS and LFM2.5 Audio 1.5B are distinct model requests." + "reason": "Different models/tasks: Qwen3-TTS vs LFM2.5 Audio 1.5B." }, { "left": "issue:43837", "right": "issue:43909", "accept": false, - "reason": "Qwen3-ASR and LFM2.5 Audio are different speech/audio models." + "reason": "Different models/tasks: Qwen3-ASR vs LFM2.5 Audio 1.5B." }, { "left": "issue:43837", "right": "issue:45322", "accept": false, - "reason": "Qwen3-ASR and EUPE target different architectures and tasks." + "reason": "Different models/tasks: Qwen3-ASR vs EUPE." }, { "left": "issue:43837", "right": "issue:44121", "accept": false, - "reason": "Qwen3-ASR support is unrelated to sparse transformer support." + "reason": "Different models/tasks: Qwen3-ASR vs circuit-sparsity transformer." }, { "left": "issue:43909", "right": "issue:44205", "accept": false, - "reason": "LFM2.5 Audio and SAM3-LiteText are different model requests." + "reason": "Different models/tasks: LFM2.5 Audio 1.5B vs SAM3-LiteText." }, { "left": "issue:44121", "right": "issue:45322", "accept": false, - "reason": "Weight-sparse transformer request vs EUPE request; not same problem." + "reason": "Different models: circuit-sparsity transformer vs EUPE." }, { "left": "issue:43837", "right": "issue:45027", "accept": false, - "reason": "Qwen3-ASR and Voxtral TTS are separate speech model support requests." + "reason": "Different models/tasks: Qwen3-ASR vs Voxtral TTS." }, { "left": "issue:37015", "right": "issue:45027", "accept": false, - "reason": "NeoBERT and Voxtral TTS are unrelated." + "reason": "Different models/tasks: NeoBERT vs Voxtral TTS." }, { "left": "issue:43837", "right": "issue:45177", "accept": false, - "reason": "Qwen3-ASR and DeepSeek-OCR2 are different model domains." + "reason": "Different models/tasks: Qwen3-ASR vs DeepSeek-OCR2." }, { "left": "issue:45322", "right": "issue:46133", "accept": false, - "reason": "EUPE and TIPSv2 are distinct vision-language model requests." + "reason": "Different model requests: EUPE vs TIPSv2. Same general domain, not the same change." } ] }, @@ -16395,12 +16446,12 @@ "issue:43717|issue:46144", "issue:46143|issue:46144" ], - "prepared_review_unit_hash": "a67a18593df395c58bc2099853d51ed0232616a085550b333ac0342a1ffccf5c", + "prepared_review_unit_hash": "86c9244c776f418640b0841225da650df5d637bd14315c4f716f948f91ade438", "component_budget": { "node_count": 10, "item_count": 10, "soft_pair_count": 13, - "serialized_chars": 10293, + "serialized_chars": 10295, "estimated_input_tokens": 2574, "estimated_eval_tokens": 5404 }, @@ -16408,7 +16459,7 @@ "node_count": 10, "item_count": 10, "soft_pair_count": 13, - "serialized_chars": 10293, + "serialized_chars": 10295, "estimated_input_tokens": 2574, "estimated_eval_tokens": 5404 }, @@ -16417,84 +16468,90 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly heterogeneous: the issues span unrelated models and bug types. The only near-related pair is the RoFormer duo, but they still describe different code paths, so I would not merge any pair as duplicates.", - "confidence": 0.9, - "canonical_issue_reason": "No clear canonical issue: the cluster is too mixed, and the only plausible relation (46143/46144) is related but not duplicate.", + "summary": "The cluster appears to be a set of unrelated model/config/documentation bugs that only share broad library vocabulary; none of the soft pairs look like true duplicates or mergeable into one fix.", + "confidence": 0.93, + "canonical_issue_reason": "If a single issue must represent the set, issue 46144 is the most concrete and best-supported report, with the strongest discussion/inbound activity and the clearest actionable code-path bug; the others are separate problems rather than duplicates.", "canonical_pr_reason": null, - "best_issue_reason": "If forced to pick one representative, issue 46144 is the closest hub because it is the broader RoFormer attention-path report and the only item with a near-neighbor, but it still should not serve as a duplicate anchor.", + "best_issue_reason": "Issue 46144 is the strongest standalone candidate: it describes a specific RoFormer attention-interface defect, has some external attention, and is more actionable than the other reports in this group.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:37934", "right": "issue:43668", "accept": false, - "reason": "Same general normalization vocabulary, but different models and different bugs: RMS-norm question vs. incorrect type hint." + "reason": "Different model components and different bug types: Llama4 norm-vs-RMS question vs a ModernBERT config type-hint issue." }, { "left": "issue:42936", "right": "issue:43240", "accept": false, - "reason": "Mask2Former ignore_value handling and loss-kwargs forwarding are different code-path issues." + "reason": "Mask2former ignore_value usage and loss-function kwargs forwarding are separate code paths and separate fixes." }, { "left": "issue:42936", "right": "issue:46018", "accept": false, - "reason": "Different model families and different behaviors; no shared concrete bug." + "reason": "Unrelated model behaviors: Mask2former loss handling vs DeepSeek-V4 shared expert gating." }, { "left": "issue:42936", "right": "issue:46144", "accept": false, - "reason": "Mask2Former loss/ignore_value bug is unrelated to RoFormer attention-interface behavior." + "reason": "No shared concrete bug; Mask2former ignore_value and RoFormer attention-interface behavior are unrelated." }, { "left": "issue:42936", "right": "issue:44031", "accept": false, - "reason": "Tokenizer regex warning and Mask2Former ignore_value are unrelated." + "reason": "Model loss-parameter issue vs tokenizer regex warning; different subsystems and fixes." }, { "left": "issue:42936", "right": "issue:43813", "accept": false, - "reason": "A typo in a quantization attribute is unrelated to Mask2Former ignore_value usage." + "reason": "Different concerns entirely: Mask2former loss usage vs a quantization-string typo." }, { "left": "issue:43240", "right": "issue:46144", "accept": false, - "reason": "Generic loss-kwargs forwarding and RoFormer attention-interface usage are different bugs in different subsystems." + "reason": "Both mention kwargs, but one is about loss-function forwarding and the other about RoFormer methods/attention wiring; not the same bug." }, { "left": "issue:43240", "right": "issue:43668", "accept": false, - "reason": "Loss-kwargs forwarding and ModernBERT norm_eps typing are unrelated." + "reason": "Loss kwargs propagation and a config type-hint typo are unrelated issues." }, { "left": "issue:43668", "right": "issue:44031", "accept": false, - "reason": "Incorrect config type hint vs. tokenizer regex warning; no common bug." + "reason": "Type-hint correctness and tokenizer regex warnings are unrelated." }, { "left": "issue:43668", "right": "issue:43813", "accept": false, - "reason": "Type-hint correction and attribute typo are different fixes." + "reason": "Both are small correctness issues, but one is a config typing mistake and the other is a typo in a different code path." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2former ignore_value behavior and ModernBERT config typing are not the same underlying bug." }, { "left": "issue:43717", "right": "issue:46144", "accept": false, - "reason": "Mamba init_weights usage and RoFormer attention interface are unrelated model-specific issues." + "reason": "Mamba init_weights behavior and RoFormer attention-interface behavior are distinct model-specific bugs." }, { "left": "issue:46143", "right": "issue:46144", "accept": false, - "reason": "Both are RoFormer-related, but one is about kwargs passthrough in methods and the other is about the attention implementation not using the interface; adjacent, not the same duplicate bug." + "reason": "Closely related RoFormer reports, but one is about kwargs not being passed through methods and the other about attention not using the interface; related area, different concrete fix." } ] }, @@ -16513,8 +16570,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-41084-2", "nodes": [ @@ -16524,7 +16581,7 @@ "soft_pairs": [ "issue:41084|issue:43037" ], - "prepared_review_unit_hash": "db73774e45ba2bccb7b976805154db3971bfa95b1062601482919f72790287f2", + "prepared_review_unit_hash": "767c8dacf4bd1f2c5b26e9ebffa3981cd7b356e7c6f4efa9949f496c13e7c2ef", "component_budget": { "node_count": 2, "item_count": 2, @@ -16546,18 +16603,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues do not look like duplicates: one is about block decoding behavior, while the other is a request for DeepSeek v3.2 model support. They target different underlying changes and should remain separate.", + "summary": "The two issues are not duplicates: one requests Set Block Decoding, while the other asks for DeepSeek v3.2 support. They describe different features/problems with no shared targets or explicit linkage.", "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": "No single canonical issue fits this cluster well because the items describe different requests; if forced, neither is a clear duplicate representative of the other.", + "best_issue_reason": null, "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:41084", "right": "issue:43037", "accept": false, - "reason": "Different topics and likely different code paths: block decoding mechanics vs adding support for a specific model/version. The overlap is too superficial to treat as the same underlying issue." + "reason": "Different underlying requests: decoding behavior vs model support. No shared code path, filenames, or explicit relation indicating the same bug/change." } ] }, @@ -16576,8 +16633,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43598-2", "nodes": [ @@ -16587,12 +16644,12 @@ "soft_pairs": [ "issue:43598|issue:43600" ], - "prepared_review_unit_hash": "490735881aa000c461859610073fe234611327fd6d229a0b0a66c3da5d63e8c2", + "prepared_review_unit_hash": "3fa11e03f42da98482734e5ed7dce4d6f74543d23229ac0765455c8633ffc54b", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1582, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -16600,7 +16657,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 1582, + "serialized_chars": 1584, "estimated_input_tokens": 396, "estimated_eval_tokens": 1048 }, @@ -16609,18 +16666,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are related only at a high level (Trainer customization), but they describe different problems: one is about `remove_unused_column` behavior, the other about FLOPS estimation for custom trainers. They should not be merged as duplicates.", - "confidence": 0.93, - "canonical_issue_reason": null, + "summary": "These look like two separate Trainer enhancement issues: one about making `remove_unused_column` more customizable, and the other about correctly estimating FLOPS for custom trainers. They share the Trainer/custom-trainer theme but not the same bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #43598 is the cleaner canonical issue for its own topic because it is a focused Trainer customization request with a clear, specific hook (`remove_unused_column`).", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43598 is the broader Trainer customizability discussion, so it is the better representative issue for this small cluster, but it is still distinct from 43600.", + "best_issue_reason": "Neither issue is a duplicate of the other, so there is no global best issue for the cluster; if forced to pick one representative, #43598 is more concrete and narrowly scoped.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43598", "right": "issue:43600", "accept": false, - "reason": "Different underlying bugs: customizability of unused-column removal vs. FLOPS accounting for custom trainers. Shared Trainer context is too broad to treat as the same issue." + "reason": "Both involve custom trainers in Trainer, but they address different underlying problems: one is about unused-column removal customization, the other about FLOPS estimation. They are not the same bug/change and should not be merged as duplicates." } ] }, @@ -16639,8 +16696,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43684-2", "nodes": [ @@ -16650,7 +16707,7 @@ "soft_pairs": [ "issue:43684|issue:44405" ], - "prepared_review_unit_hash": "16b971e76369779cd8e72b8063b02eda11eed670163d9ac6a80cc9cf0d7ae5db", + "prepared_review_unit_hash": "66e707e517b28e39997d8cd378a0b3b3a062cf4f755ca00b34bfa2183808a516", "component_budget": { "node_count": 2, "item_count": 2, @@ -16672,18 +16729,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related only at a very broad Transformers-auto-model level, but they target different model families and different auto classes, so they are not duplicates.", + "summary": "These are not duplicates: one requests Qwen3-Omni registration in generic auto-model classes, while the other asks for sequence-classification support for Qwen3.5. They involve different model families and different API surfaces/tasks.", "confidence": 0.96, - "canonical_issue_reason": "issue:43684 is the cleaner canonical issue because it is the more direct registration request and names the exact auto classes involved.", + "canonical_issue_reason": "Issue 43684 is the broader, more central registry gap request and is the better representative of a model-registration problem.", "canonical_pr_reason": null, - "best_issue_reason": "issue:43684 is the best representative since it is a concrete API registration gap; issue:44405 is a separate Qwen3.5 sequence-classification support request.", + "best_issue_reason": "Issue 43684 is the clearest standalone issue to keep as the canonical issue because it describes a general AutoModel/AutoModelForConditionalGeneration registration addition.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43684", "right": "issue:44405", "accept": false, - "reason": "Different underlying change requests: Qwen3-Omni auto-registration for AutoModel/AutoModelForConditionalGeneration vs Qwen3.5 AutoModelForSequenceClassification support. They touch different model families and different code paths, so they should not be merged as duplicates." + "reason": "Different underlying change request: Qwen3-Omni auto-model registration vs Qwen3.5 sequence-classification support. Different model family and different code path." } ] }, @@ -16702,8 +16759,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-43835-4", "nodes": [ @@ -16720,77 +16777,80 @@ "issue:43845|issue:43859", "issue:43846|issue:43859" ], - "prepared_review_unit_hash": "3204fc7f6f266e3645e342519ff09205bc61722498f3bd6effe15ea6b39fd45b", + "prepared_review_unit_hash": "32c7832b1cbeb78151b3a483f4355d28cf5b7cc3d81c6bec504b425ff90fcc3a", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4261, - "estimated_input_tokens": 1066, - "estimated_eval_tokens": 2388 + "serialized_chars": 4259, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 4261, - "estimated_input_tokens": 1066, - "estimated_eval_tokens": 2388 + "serialized_chars": 4259, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is mostly low-signal spam-like issue reports. Three issues share the same generic title \u201chuggingface \u201d and are likely duplicates of each other, while \u201cFett\u201d looks unrelated. No PRs are present.", - "confidence": 0.39, - "canonical_issue_reason": "Issue 43845 is the best available anchor because it is the earliest of the identical \u201chuggingface \u201d reports, but the bodies are too sparse to identify a concrete underlying bug with confidence.", + "summary": "The only strong duplicate signal is the near-simultaneous pair 43845/43846, both titled \"huggingface \". 43859 shares the same title but is separated by many hours, so it is a weaker match and should not be merged on title alone. 43835 (\"Fett\") is unrelated.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 43845 is the best canonical representative for the clear duplicate pair because it is the earlier of the two near-identical \"huggingface \" reports.", "canonical_pr_reason": null, - "best_issue_reason": "Issue 43845 is the strongest cluster representative: it matches the repeated title pattern, is earliest among the duplicate-looking reports, and there is no stronger technical signal from the bodies.", + "best_issue_reason": "43845 is the clearest representative of the duplicate set: it matches 43846 closely and is the earliest of that tight duplicate pair.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:43835", "right": "issue:43845", "accept": false, - "reason": "Different titles and no body detail; only weak lexical similarity, not enough to infer the same issue." + "reason": "\"Fett\" is unrelated to the repeated \"huggingface \" reports; no evidence they describe the same issue." }, { "left": "issue:43835", "right": "issue:43846", "accept": false, - "reason": "Different titles and no substantive description; superficial similarity only." + "reason": "\"Fett\" is unrelated to the repeated \"huggingface \" reports; no evidence they describe the same issue." }, { "left": "issue:43835", "right": "issue:43859", "accept": false, - "reason": "Different titles and no concrete evidence of a shared underlying bug." + "reason": "\"Fett\" is unrelated to the repeated \"huggingface \" reports; no evidence they describe the same issue." }, { "left": "issue:43845", "right": "issue:43846", - "accept": false, - "reason": "Same generic title, but the reports are too empty to confirm they describe the same concrete problem." + "accept": true, + "reason": "Same title and created within minutes, which strongly suggests the same underlying duplicate issue." }, { "left": "issue:43845", "right": "issue:43859", "accept": false, - "reason": "Both are generic \u201chuggingface \u201d reports, but there is no actionable description tying them to the same underlying issue." + "reason": "Same title, but the reports are separated by many hours and there is no additional evidence beyond soft similarity to confirm the same underlying issue." }, { "left": "issue:43846", "right": "issue:43859", "accept": false, - "reason": "Identical superficial title, but insufficient detail to safely treat them as duplicates of the same bug." + "reason": "Same title, but the reports are separated by many hours and there is no additional evidence beyond soft similarity to confirm the same underlying issue." } ] }, - "evaluator_result": null, - "evaluator_used": false, - "retried": false, - "accepted_nontrivial_soft_edge": false, + "evaluator_result": { + "accept": false, + "feedback": "The grouping is mostly grounded, but the soft-edge reasoning is not fully conservative. In particular, the 43845/43859 and 43846/43859 pairs are separated by many hours, so timing does not really support a duplicate inference there; the evidence is only the identical title/soft similarity. The summary and canonical choice for 43845 are reasonable, but the duplicate verdicts should be phrased more cautiously." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null }, @@ -16802,8 +16862,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-44371-5", "nodes": [ @@ -16824,88 +16884,88 @@ "issue:44572|issue:44937", "issue:44573|issue:44937" ], - "prepared_review_unit_hash": "89aa2061b3170f8941b6029cfe407813b221f7e8f1b123ab9358e7b8b707a4c6", + "prepared_review_unit_hash": "310c6b85729a16148216edff379f05ec32b7082f4305b8de3f59c4372f771e11", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5785, - "estimated_input_tokens": 1447, - "estimated_eval_tokens": 3150 + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 9, - "serialized_chars": 5785, - "estimated_input_tokens": 1447, - "estimated_eval_tokens": 3150 + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is made up of low-information spam/generic issue titles with no substantive description, links, or code evidence. There is no defensible duplicate grouping or canonical issue here.", - "confidence": 0.96, - "canonical_issue_reason": null, + "summary": "The cluster is dominated by spammy or content-free issues with only one mildly descriptive title, and there is no evidence that any pair is the same underlying bug or request. Treat them as separate low-signal issues rather than duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #44418 is the least placeholder-like item because it has a descriptive title, but it still lacks any concrete problem statement and does not match the others as a clear duplicate.", "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "#44418 is the best representative issue only because its title is more specific than the pure spam entries; it is not a strong canonical duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:44371", "right": "issue:44572", "accept": false, - "reason": "Both are titled '', but there is no body, target, or issue-specific evidence to show they describe the same underlying problem." + "reason": "Both are just placeholder spam titles with no substantive body or issue details; no concrete shared bug or request is visible." }, { "left": "issue:44371", "right": "issue:44573", "accept": false, - "reason": "Both look like spam placeholders only; no concrete bug/change overlap is visible." + "reason": "These appear to be separate spam entries rather than the same issue, with no identifying context tying them together." }, { "left": "issue:44371", "right": "issue:44418", "accept": false, - "reason": "One is spam-like and the other is a generic documentation suggestion; insufficient evidence they are the same issue." + "reason": "One is placeholder spam and the other is a generic documentation suggestion; too little evidence that they describe the same underlying request." }, { "left": "issue:44371", "right": "issue:44937", "accept": false, - "reason": "Titles are unrelated ('' vs 'Check out \"Google LLC\"'); no sign of the same underlying report." + "reason": "'Check out \"Google LLC\"' does not align with the spam placeholder title as the same underlying issue." }, { "left": "issue:44418", "right": "issue:44572", "accept": false, - "reason": "A generic documentation suggestion does not match a spam placeholder issue in a way that supports duplication." + "reason": "A generic documentation enhancement and a spam placeholder are not enough to conclude a duplicate relationship." }, { "left": "issue:44418", "right": "issue:44573", "accept": false, - "reason": "No concrete shared bug or request is evident beyond being low-information issues." + "reason": "No concrete overlap in problem statement or requested change; these look like unrelated low-quality issues." }, { "left": "issue:44572", "right": "issue:44573", "accept": false, - "reason": "Identical spam titles are not enough to confirm the same underlying issue without any descriptive evidence." + "reason": "Although both are placeholder titles, that alone is not enough to merge them as the same underlying issue." }, { "left": "issue:44572", "right": "issue:44937", "accept": false, - "reason": "Spam placeholder and unrelated promotional-style title; no common issue is visible." + "reason": "The titles suggest different spam/content themes, not a single shared bug or request." }, { "left": "issue:44573", "right": "issue:44937", "accept": false, - "reason": "No substantive overlap in the report content or title." + "reason": "No shared concrete problem description or target is present; these should not be deduplicated." } ] }, @@ -16924,8 +16984,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45145-2", "nodes": [ @@ -16935,7 +16995,7 @@ "soft_pairs": [ "issue:45145|issue:45151" ], - "prepared_review_unit_hash": "1c153ba80fbaa93ada231663e59220cc66d4145b222f97b318b37bc1ac73e66e", + "prepared_review_unit_hash": "00ab2e5e779fc12cf48acc5fc2b3c6c483431e5917284572e8ab1f70f8ec80e7", "component_budget": { "node_count": 2, "item_count": 2, @@ -16957,24 +17017,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster appears to be a near-duplicate issue pair: same title, close similarity signal, and no contrary evidence. However, the evidence is not strong enough to claim identical bodies or a confirmed duplicate relationship with certainty.", - "confidence": 0.84, - "canonical_issue_reason": "Issue #45145 is the best canonical issue because it is the earlier thread and has an inbound reference, while the pair only shows soft similarity rather than explicit duplicate linkage.", + "summary": "The two closed issues are highly similar and likely related duplicate reports, but the packet only establishes title-level similarity and does not prove they are the same underlying proposal. Issue #45145 is the better anchor because it is earlier and already has an inbound reference.", + "confidence": 0.82, + "canonical_issue_reason": "Issue #45145 is the better canonical issue because it was created first and already has an inbound reference, so it is the likelier discussion anchor for a probable duplicate pair.", "canonical_pr_reason": null, - "best_issue_reason": "#45145 is the strongest representative for the issue cluster: same title as #45151, slightly earlier creation time, and it has the only inbound reference in the pair.", + "best_issue_reason": "Issue #45145 is the best global issue candidate for the same reason: it is the earliest and most referenced of the two near-identical reports.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45145", "right": "issue:45151", "accept": false, - "reason": "The titles match and the similarity score is fairly high, but the packet only provides soft-similarity evidence and marks deterministic_accept=false. Without body text or explicit links, this is not strong enough to confirm a duplicate." + "reason": "The titles are identical and the reports look very similar, but the evidence is not strong enough to confirm they are the same underlying issue. There are also small distinguishing signals, including inbound reference and activity/body-length differences." } ] }, "evaluator_result": { "accept": false, - "feedback": "The packet supports a close similarity, but not the stronger claims made here. Both issues have the same title and similar body lengths, with only a soft-similarity signal; the bodies themselves are not shown, so saying they have the \"same title/body\" or are \"effectively the same issue content\" overstates the evidence. Also, the soft-edge verdict should stay conservative because the packet marks this pair as deterministic_accept=false. The inbound reference on 45145 is fine to mention, but it does not by itself justify treating the pair as confirmed duplicates." + "feedback": "The duplicate call is directionally plausible, but the summary overstates what the packet proves. The packet supports strong title similarity and a likely duplicate relationship, yet it does not establish that the two issues are the same underlying proposal, and it is not true that there is \u201cno distinguishing metadata beyond creation time\u201d because issue #45145 has an inbound reference while #45151 does not, and their discussion activity/body length also differ. Please soften the language to match the evidence." }, "evaluator_used": true, "retried": true, @@ -17044,8 +17104,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45672-2", "nodes": [ @@ -17055,7 +17115,7 @@ "soft_pairs": [ "issue:45672|issue:45674" ], - "prepared_review_unit_hash": "56d439d7b9e3449ae8a252d2e76128c11d02c20a04bbf7bd11e58264b071f025", + "prepared_review_unit_hash": "feab38ee0466aeb63912ce8f83c339fabd9ce085cb7cbb9572a02af7fe9ea9ce", "component_budget": { "node_count": 2, "item_count": 2, @@ -17077,7 +17137,7 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues are both about 4-bit inference, but they describe different bugs: one is a Gemma4 audio-module uint8/torch.finfo TypeError, the other is a BitsAndBytesConfig regression where llm_int8_skip_modules drops the default lm_head exclusion and triggers an AssertionError.", + "summary": "These are related to 4-bit inference, but they describe different failures with different triggers and code paths, so they should not be merged as duplicates.", "confidence": 0.96, "canonical_issue_reason": null, "canonical_pr_reason": null, @@ -17088,7 +17148,7 @@ "left": "issue:45672", "right": "issue:45674", "accept": false, - "reason": "Reject: different root causes, different affected code paths, and different failure modes; they are not the same underlying bug." + "reason": "Different bugs: one is a Gemma4/audio-module torch.finfo() TypeError on uint8 weights during NF4 inference, the other is an AssertionError caused by llm_int8_skip_modules clearing the default lm_head exclusion. Same broad area, but not the same underlying issue." } ] }, @@ -17107,8 +17167,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45762-2", "nodes": [ @@ -17118,7 +17178,7 @@ "soft_pairs": [ "issue:45762|issue:45763" ], - "prepared_review_unit_hash": "0c2e3cb79c508a9df230e546276fcb842eee53984e12c91caa815a680b38678b", + "prepared_review_unit_hash": "c1a394bda8158756b4a565994cc44969a493062fa34978f9995c3b4d81ae4212", "component_budget": { "node_count": 2, "item_count": 2, @@ -17140,8 +17200,8 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two closed issues look like unrelated personal-name entries with different names and dates; there is no evidence they describe the same underlying bug or request.", - "confidence": 0.98, + "summary": "These are two closed issues with unrelated person-name/date titles and no body or linking evidence, so they do not look like duplicates or the same underlying report.", + "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, @@ -17151,7 +17211,7 @@ "left": "issue:45762", "right": "issue:45763", "accept": false, - "reason": "Same superficial pattern (name + date), but the names and dates differ and there is no shared bug report, target, or context indicating a duplicate." + "reason": "Different named individuals and dates, with no shared description, references, or code context; not the same underlying issue." } ] }, @@ -17170,8 +17230,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45764-2", "nodes": [ @@ -17181,7 +17241,7 @@ "soft_pairs": [ "issue:45764|issue:45765" ], - "prepared_review_unit_hash": "65495026d540e24b9a98a2d521580d863f19902394227820a17dff3192a27703", + "prepared_review_unit_hash": "7d379b7281f27058ab2bf9bdadb60710273c675e5fa391a1d28e7f874d03b264", "component_budget": { "node_count": 2, "item_count": 2, @@ -17203,18 +17263,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two issues appear unrelated: they reference different personal names and different dates, with no body or links suggesting a shared bug or task. The soft duplicate edge should be rejected.", - "confidence": 0.98, + "summary": "The two issues only share a loose title pattern (a person\u2019s name plus date) and provide no body, technical context, or linkage showing they refer to the same underlying bug/task. They should not be clustered as duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, "canonical_pr_reason": null, - "best_issue_reason": null, + "best_issue_reason": "No strong canonical issue candidate: both are sparse, non-technical entries with different names and dates, so neither can be confidently treated as the duplicate target.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45764", "right": "issue:45765", "accept": false, - "reason": "Different names and different dates; no evidence they describe the same underlying issue." + "reason": "Reject: the titles refer to different people and different dates, with no body or explicit linkage indicating the same underlying issue." } ] }, @@ -17233,8 +17293,8 @@ "review_unit_count": 1, "status": "reviewed", "reason": null, - "source": "cache", - "cache_hit": true, + "source": "llm", + "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", "cluster_id": "cluster-45869-5", "nodes": [ @@ -17254,12 +17314,12 @@ "issue:45871|issue:45873", "issue:45872|issue:45873" ], - "prepared_review_unit_hash": "7e698274b5264df30ca0ae56a37bc5e9ad7c488e4d71fa45821843f61ba777c6", + "prepared_review_unit_hash": "55d255125c68e555ba3adc3ca72b51c2279bbf06a28ea1aee3af376143d1559e", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5728, + "serialized_chars": 5726, "estimated_input_tokens": 1432, "estimated_eval_tokens": 3120 }, @@ -17267,7 +17327,7 @@ "node_count": 5, "item_count": 5, "soft_pair_count": 8, - "serialized_chars": 5728, + "serialized_chars": 5726, "estimated_input_tokens": 1432, "estimated_eval_tokens": 3120 }, @@ -17276,69 +17336,69 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two exact duplicate pairs are present: 45870/45871 and 45872/45873. Issue 45869 is related by topic but is a different dataset-style post, and the two duplicate pairs are not the same underlying issue.", - "confidence": 0.93, - "canonical_issue_reason": "No single canonical issue spans the whole set; the cluster splits into two separate duplicate pairs. If forced to pick one representative, 45870 is the clearest canonical for the 'Example: Financial time series prediction with DataForge' pair.", + "summary": "Two issue pairs look highly similar and are likely duplicates: 45870/45871 and 45872/45873. The packet only shows titles and body lengths, so this should be treated as strong duplicate evidence rather than a confirmed exact match. Issue 45869 is a separate topic.", + "confidence": 0.91, + "canonical_issue_reason": "45870 is the earliest item in the duplicated \"Example: Financial time series prediction with DataForge\" pair and is the cleanest representative of that issue thread; 45869 is unrelated, and 45872/45873 form a separate duplicate pair.", "canonical_pr_reason": null, - "best_issue_reason": "45870 is a strong issue representative because it is a clean, duplicate-bearing copy of the 'Example: Financial time series prediction with DataForge' report and is simpler than the broader, less clearly comparable items.", + "best_issue_reason": "45870 is the best overall issue representative because it is the earliest clear duplicate in the packet and has the most straightforward canonical title among the closely matching items.", "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "issue:45869", "right": "issue:45870", "accept": false, - "reason": "Same broad finance theme, but one is a dataset promo and the other is an example post; not the same underlying issue." + "reason": "Different topics: a dataset bundle listing versus a financial time series prediction example." }, { "left": "issue:45869", "right": "issue:45871", "accept": false, - "reason": "Same broad finance theme, but different artifact types and wording; not a duplicate." + "reason": "Different topics: a dataset bundle listing versus a financial time series prediction example." }, { "left": "issue:45870", "right": "issue:45871", "accept": true, - "reason": "Identical title and matching body size strongly indicate the same issue duplicated." + "reason": "Same title and matching body length; very likely the same issue content, though the full bodies are not shown." }, { "left": "issue:45870", "right": "issue:45872", "accept": false, - "reason": "Related DataForge/financial wording, but example post vs scraping service are different underlying topics." + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." }, { "left": "issue:45870", "right": "issue:45873", "accept": false, - "reason": "Related wording only; the two issues describe different concepts and are not the same bug/change." + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." }, { "left": "issue:45871", "right": "issue:45872", "accept": false, - "reason": "Similar domain language, but no evidence they report the same underlying issue." + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." }, { "left": "issue:45871", "right": "issue:45873", "accept": false, - "reason": "Similar domain language, but these are different issue subjects." + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." }, { "left": "issue:45872", "right": "issue:45873", "accept": true, - "reason": "Identical title and matching body size strongly indicate the same issue duplicated." + "reason": "Same title and matching body length; very likely the same issue content, though the full bodies are not shown." } ] }, "evaluator_result": { - "accept": true, - "feedback": "The summary is well grounded in the packet: the only strongly supported duplicate pairs are 45870/45871 and 45872/45873 (matching titles and identical body lengths), while 45869 is only loosely related by theme. The soft-edge verdicts are conservative and do not overclaim duplicates beyond the evidence provided." + "accept": false, + "feedback": "The summary is a bit too strong for the packet. The evidence supports that 45870/45871 and 45872/45873 are highly similar and likely duplicates (identical titles and matching body lengths), but the packet does not show the full bodies, so calling them \u201cexact duplicates\u201d overstates certainty. The statement that 45869 is a separate topic is reasonable, but the duplicate verdicts should be phrased more conservatively." }, "evaluator_used": true, - "retried": false, + "retried": true, "accepted_nontrivial_soft_edge": true, "error_kind": null, "error_message": null @@ -17395,172 +17455,178 @@ "pull_request:43823|pull_request:45213", "pull_request:43251|pull_request:43254" ], - "prepared_review_unit_hash": "90b99c246b839d284738ce6a3c3bd73887f93bfffa2378cfea484a2998a57b94", + "prepared_review_unit_hash": "f3e9fdc8479def9b7ae00e8aebac33c70ce931040c8143465ce83af123d1afd6", "component_budget": { "node_count": 14, "item_count": 14, "soft_pair_count": 22, - "serialized_chars": 38371, - "estimated_input_tokens": 9593, - "estimated_eval_tokens": 19442 + "serialized_chars": 38366, + "estimated_input_tokens": 9592, + "estimated_eval_tokens": 19440 }, "budget": { "node_count": 14, "item_count": 14, "soft_pair_count": 22, - "serialized_chars": 38371, - "estimated_input_tokens": 9593, - "estimated_eval_tokens": 19442 + "serialized_chars": 38366, + "estimated_input_tokens": 9592, + "estimated_eval_tokens": 19440 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are mostly unrelated PRs across separate areas (new models, docs, loading, tests, and utility refactors). The only true duplicate-style overlap is the pair of `fixed_cross_entropy` patches, which both implement the same kwargs support in the same function for the same tracked issue.", - "confidence": 0.91, + "summary": "The cluster is mostly unrelated PRs spanning model additions, refactors, and test/docs changes. The only true duplicate pair is the two fixed_cross_entropy PRs, which patch the same loss helper with the same kwargs support.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "`pull_request:43251` is the cleanest representative: it directly targets issue 43240 and implements the concrete `fixed_cross_entropy` kwargs extension in one place.", + "canonical_pr_reason": "pull_request:43251 is the best representative for the loss fix: it has the explicit issue link, targets the exact bug in fixed_cross_entropy, and its change is the cleanest description of the shared code-path fix.", "best_issue_reason": null, - "best_pr_reason": "`pull_request:43251` is the best PR candidate because it has the explicit issue link and the most direct, narrowly-scoped implementation of the loss fix.", + "best_pr_reason": "pull_request:43251 is the strongest overall PR in this cluster for duplicate triage because it most directly and clearly addresses the shared bug; the other near-match (43254) is just an alternate wording of the same patch.", "soft_edge_verdicts": [ { "left": "pull_request:39895", "right": "pull_request:43424", "accept": false, - "reason": "VideoPrism model addition/docs is unrelated to an executorch dynamic-shape test change." + "reason": "Unrelated PRs: VideoPrism model/docs addition vs Executorch dynamic-shape test." }, { "left": "pull_request:41855", "right": "pull_request:43823", "accept": false, - "reason": "Mistral tokenizer method additions and a new MobileLLM model are different changes in different subsystems." + "reason": "Different models and different changes: Mistral tokenizer methods vs adding MobileLLM." }, { "left": "pull_request:41855", "right": "pull_request:43424", "accept": false, - "reason": "Tokenizer API additions are unrelated to executorch export test coverage." + "reason": "Tokenizer API fix vs Executorch export test; no shared bug or code path." }, { "left": "pull_request:42668", "right": "pull_request:43823", "accept": false, - "reason": "A processor-from-pretrained robustness refactor is unrelated to adding a new model." + "reason": "Processor loading robustness changes for existing models vs adding a new model implementation." }, { "left": "pull_request:43251", "right": "pull_request:46067", "accept": false, - "reason": "`fixed_cross_entropy` kwargs support and loading/base_model_prefix conversion fixes are different code paths and problems." + "reason": "Loss helper kwargs fix vs conversion/loading prefix handling; different subsystems and bugs." }, { "left": "pull_request:43251", "right": "pull_request:43823", "accept": false, - "reason": "Loss-function kwargs support is unrelated to a new model implementation." + "reason": "Cross-entropy helper fix vs MobileLLM addition; unrelated." }, { "left": "pull_request:43251", "right": "pull_request:44827", "accept": false, - "reason": "A loss utility fix is unrelated to Mistral4 test/debug changes." + "reason": "Loss helper kwargs fix vs Mistral4 test/model adjustments; different issues." }, { "left": "pull_request:43424", "right": "pull_request:46067", "accept": false, - "reason": "Executorch dynamic-shape tests do not match the conversion/loading prefix fix." + "reason": "Executorch dynamic-shape test vs base_model_prefix conversion fix; unrelated." }, { "left": "pull_request:43424", "right": "pull_request:44827", "accept": false, - "reason": "Executorch export tests and Mistral4 test fixes are separate areas with no shared underlying bug." + "reason": "Executorch export testing vs Mistral4 test/model changes; not the same bug." }, { "left": "pull_request:43424", "right": "pull_request:43823", "accept": false, - "reason": "Executorch exportability testing is unrelated to adding MobileLLM." + "reason": "Executorch export test vs new MobileLLM model; unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45655", "accept": false, - "reason": "A new model addition is unrelated to fixing config resolution order in modeling_utils." + "reason": "New model implementation vs config resolution order fix; no shared concrete change." }, { "left": "pull_request:43823", "right": "pull_request:43995", "accept": false, - "reason": "MobileLLM implementation and Falcon output-collection refactoring are unrelated model changes." + "reason": "New model addition vs Falcon output-interface refactor; different targets." }, { "left": "pull_request:43823", "right": "pull_request:46067", "accept": false, - "reason": "New MobileLLM files and base_model_prefix loading/conversion fixes do not address the same bug." + "reason": "New model addition vs conversion prefix-loading fix; unrelated." }, { "left": "pull_request:43995", "right": "pull_request:44827", "accept": false, - "reason": "Falcon output-collection refactor and Mistral4 test fixes are different changes." + "reason": "Falcon output-capture refactor vs Mistral4 fixes; different code paths." }, { "left": "pull_request:43995", "right": "pull_request:45213", "accept": false, - "reason": "A model refactor is unrelated to the model-creation skill/meta change." + "reason": "Falcon refactor vs model-creation skill/meta PR; not a duplicate bug fix." }, { "left": "pull_request:44827", "right": "pull_request:46067", "accept": false, - "reason": "Mistral4 test fixes and conversion base_model_prefix handling are unrelated." + "reason": "Mistral4 test/model changes vs conversion base_model_prefix loading fix; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45170", "accept": false, - "reason": "Mistral4 test fixes and `layrnorm`\u2192`layernorm` renaming are different bugs." + "reason": "Mistral4 fixes vs layrnorm/layernorm rename cleanup; different underlying issue." }, { "left": "pull_request:44827", "right": "pull_request:46048", "accept": false, - "reason": "Mistral4 tests are unrelated to adding a TDT loss kernel." + "reason": "Mistral4 fixes vs TDT loss kernel addition; unrelated." }, { "left": "pull_request:44827", "right": "pull_request:45655", "accept": false, - "reason": "Mistral4 test cleanup does not overlap with config resolution order fix." + "reason": "Mistral4 fixes vs config resolution order fix; not the same change." }, { "left": "pull_request:44827", "right": "pull_request:45213", "accept": false, - "reason": "A model test fix and a repository skill/meta PR are not the same change." + "reason": "Mistral4-related code vs model-creation skill/meta PR; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:43254", + "accept": false, + "reason": "Mistral4 fixes vs cross_entropy kwargs patch; unrelated." }, { "left": "pull_request:43823", "right": "pull_request:45213", "accept": false, - "reason": "Adding MobileLLM is unrelated to the model-creation skill PR." + "reason": "New MobileLLM model vs meta skill PR; not a duplicate." }, { "left": "pull_request:43251", "right": "pull_request:43254", "accept": true, - "reason": "Both PRs modify `fixed_cross_entropy` to add support for cross-entropy kwargs/parameters for the same tracked issue, and they plausibly collapse into one implementation." + "reason": "Both PRs fix the same fixed_cross_entropy helper to forward cross_entropy kwargs on the same code path, and could be merged into a single change." } ] }, "evaluator_result": { "accept": true, - "feedback": "The summary is grounded in the packet and stays conservative on the soft edges. The only accepted pair is well-supported: both PRs touch `src/transformers/loss/loss_utils.py`, target issue 43240, and appear mergeable into a single kwargs-support fix for `fixed_cross_entropy`. The rejected pairs are also reasonably treated as unrelated or too distant to be the same bugfix." + "feedback": "The summary is grounded in the packet, and the soft-edge verdicts stay conservative. The only accepted pair, pull_request:43251 with pull_request:43254, is clearly mergeable into one PR for the same fixed_cross_entropy kwargs bugfix. The rest are reasonable non-duplicate rejections based on distinct subsystems, targets, or change types." }, "evaluator_used": true, "retried": false, @@ -17587,40 +17653,40 @@ "soft_pairs": [ "pull_request:40546|pull_request:42781" ], - "prepared_review_unit_hash": "4dafadc9eb5ba9c160cf3132ab6aa5d7d253c32557d466de11ba8a7ff5fd1c74", + "prepared_review_unit_hash": "b7ebb48bdff3221cfad0c0410f23a71d3907c42ae57a85242833cf66cef027bf", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 7226, - "estimated_input_tokens": 1807, - "estimated_eval_tokens": 3870 + "serialized_chars": 7224, + "estimated_input_tokens": 1806, + "estimated_eval_tokens": 3868 }, "budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 7226, - "estimated_input_tokens": 1807, - "estimated_eval_tokens": 3870 + "serialized_chars": 7224, + "estimated_input_tokens": 1806, + "estimated_eval_tokens": 3868 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two related VibeVoice pull requests, but they target different scope: one is the initial VibeVoice integration, the other is a newer VibeVoice Realtime/acoustic-tokenizer extension. They overlap on shared model infrastructure, but are not the same change.", - "confidence": 0.91, + "summary": "These are related VibeVoice pull requests, but they are not the same change: one implements the base VibeVoice integration, while the other adds a separate realtime variant plus acoustic tokenizer work. The overlap is mostly shared model scaffolding, not a single duplicate fix or mergeable one-PR patch.", + "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "PR #42781 is the broader, newer VibeVoice effort and is the better representative of the cluster because it includes the realtime and acoustic-tokenizer work in addition to the base model integration.", + "canonical_pr_reason": "PR #40546 is the main baseline VibeVoice implementation and has substantial review activity; #42781 is a draft realtime extension with additional tokenizer/generation changes, so it is not the same concrete change.", "best_issue_reason": null, - "best_pr_reason": "PR #42781 is the strongest single anchor for this cluster: it is newer, more complete, and covers the latest VibeVoice-related implementation area, while #40546 is an earlier base implementation.", + "best_pr_reason": "PR #40546 is the best representative PR for the cluster because it is the primary full VibeVoice integration and is more mature than the draft realtime follow-up.", "soft_edge_verdicts": [ { "left": "pull_request:40546", "right": "pull_request:42781", "accept": false, - "reason": "Same model family, but not the same concrete change. #40546 implements base VibeVoice support, while #42781 adds VibeVoice Realtime plus acoustic-tokenizer pieces and extra generation changes. The overlap is broad infrastructure, not a duplicate PR that could simply be merged into one." + "reason": "Shared filenames show common VibeVoice groundwork, but the titles and diff scope differ: base VibeVoice implementation versus realtime-specific additions with acoustic tokenizer support. They are related, not the same underlying change, and would not plausibly be merged into one PR without conflating distinct features." } ] }, @@ -17650,7 +17716,7 @@ "soft_pairs": [ "pull_request:40755|pull_request:41224" ], - "prepared_review_unit_hash": "6b495204b3aa97a010a0c860b253f1eba78fd8fa67db76ca31717a28d77e1c20", + "prepared_review_unit_hash": "8c869560bef32a5fbd2d7ceb4a4fa3968c80e790cd5b73dbc241fb0f48f25384", "component_budget": { "node_count": 2, "item_count": 2, @@ -17672,18 +17738,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "False positive cluster: one PR adds TimesFM covariate forecasting, the other adds DINOv3 image-classification support. They target different model families and code paths.", + "summary": "These are unrelated pull requests: one adds TimesFM covariate forecasting, the other adds DINOv3 image-classification support and docs. They do not fix the same bug or implement the same change.", "confidence": 0.99, "canonical_issue_reason": null, - "canonical_pr_reason": "No true duplicate exists; if a representative PR is needed, 41224 is the cleaner, more self-contained feature addition.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "41224 is the more focused and review-backed standalone change, making it a better cluster representative than 40755.", + "best_pr_reason": "No single best PR for duplication triage; the PRs target different models, code paths, and user-facing features.", "soft_edge_verdicts": [ { "left": "pull_request:40755", "right": "pull_request:41224", "accept": false, - "reason": "These are unrelated features in different subsystems (TimesFM covariate forecasting vs. DINOv3 image classification) and do not fix the same concrete code-path problem." + "reason": "Different subsystems and different changes: TimesFM covariate forecasting vs DINOv3 image classification support. They are not the same underlying change." } ] }, @@ -17778,7 +17844,7 @@ "pull_request:41797|pull_request:45186", "pull_request:45186|pull_request:45630" ], - "prepared_review_unit_hash": "a98e6b55145ba8e5cb9f239c7ee81ad8979c24cab6e6b0789d6a06c0ff7dee5a", + "prepared_review_unit_hash": "9a5320ea4eb7befdef29cb035d4fd22dd6098546d88639bfc1cff8c3d8fae09f", "component_budget": { "node_count": 3, "item_count": 3, @@ -17800,24 +17866,24 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are three separate model-addition PRs (DeepSeek OCR, Isaac, and Kimi2-6). They share generic Transformers integration scaffolding, but they do not appear to fix or add the same concrete model/change.", - "confidence": 0.98, + "summary": "These are three separate model-addition PRs with overlapping scaffolding but different target models and code paths. The soft similarities are only at the shared Transformers integration pattern level, not the same underlying change.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: each PR adds a different model with its own files, tests, and model-specific implementation, so they should not be collapsed as duplicates.", + "canonical_pr_reason": "No canonical PR: the three PRs add different models (DeepSeek OCR, Isaac, Kimi2-6) and are not mergeable as one duplicate fix/change.", "best_issue_reason": null, - "best_pr_reason": "Kimi2-6 is the best anchor only in the sense that it has the highest discussion/activity and a full integration surface, but it is still a distinct model-addition PR, not a duplicate of the others.", + "best_pr_reason": "#45630 is the strongest representative of the cluster only in the sense that it is the most complete integration PR, but it is still distinct from the others and should not be treated as a duplicate.", "soft_edge_verdicts": [ { "left": "pull_request:41797", "right": "pull_request:45186", "accept": false, - "reason": "Different model additions (DeepSeek OCR vs Isaac); shared auto-registration/doc scaffolding is generic and not the same underlying change." + "reason": "Different model additions for DeepSeek OCR vs Isaac; shared auto-registration/docs patterns do not indicate the same bug or change." }, { "left": "pull_request:45186", "right": "pull_request:45630", "accept": false, - "reason": "Different model additions (Isaac vs Kimi2-6); overlap is limited to standard model onboarding files, not one concrete bug or feature." + "reason": "Different model additions for Isaac vs Kimi2-6; they share framework plumbing but fix/add separate model integrations and could not plausibly be merged into one PR." } ] }, @@ -17847,7 +17913,7 @@ "soft_pairs": [ "pull_request:42975|pull_request:42976" ], - "prepared_review_unit_hash": "bba96240a4867df568492442fe3370afcf2f4abf0a78a4ebfcac7e69ab59a36e", + "prepared_review_unit_hash": "02ef4b642c6eac7fb0ebdcfc1bf6becbdb2df91d4418a742897eccdc0919ca44", "component_budget": { "node_count": 2, "item_count": 2, @@ -17869,18 +17935,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open PRs both refresh GitHub Actions dependencies, but they target different action families and different maintenance goals: one is a Node 24 compatibility sweep, the other is a docker/build-push-action upgrade. They are related maintenance work, not the same underlying change.", - "confidence": 0.95, + "summary": "Two open PRs both update GitHub Actions workflow dependencies, but they target different upgrades: one is an umbrella Node 24 compatibility sweep, the other specifically bumps docker/build-push-action to v6. They overlap in some workflow files but do not fix the same concrete change.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 42975 is the better canonical representative because it is the broader, more clearly scoped workflow-actions compatibility update (checkout/cache/upload-artifact and related workflow runners) and its title reflects the concrete maintenance goal.", + "canonical_pr_reason": "PR 42975 is the better canonical representative because it has the broader, explicit goal of making workflows compatible with Node 24 and updates multiple core GitHub Actions references across many workflows.", "best_issue_reason": null, - "best_pr_reason": "PR 42975 is the best single PR to represent this cluster because it covers the broader GitHub Actions compatibility refresh across many workflows; PR 42976 is a narrower docker-action version bump and not the same change.", + "best_pr_reason": "PR 42975 is the best overall representative for the cluster since it captures the broader maintenance effort; PR 42976 is a narrower action-version bump focused on docker/build-push-action.", "soft_edge_verdicts": [ { "left": "pull_request:42975", "right": "pull_request:42976", "accept": false, - "reason": "Both are workflow dependency upgrades, but they touch different actions and different code paths: 42975 is about Node 24-compatible GitHub Actions versions, while 42976 upgrades docker/build-push-action. They are not the same underlying fix and would not naturally merge into one PR." + "reason": "They are not the same underlying change: 42975 upgrades several GitHub Actions for Node 24 compatibility, while 42976 only bumps docker/build-push-action versions. Shared workflow files alone are not enough to merge these as duplicates." } ] }, @@ -17917,70 +17983,70 @@ "pull_request:43448|pull_request:45490", "pull_request:43451|pull_request:45490" ], - "prepared_review_unit_hash": "661cf84cca2c042cddd20d0ab8c57948138ff93fbd14d81d110f44613c346433", + "prepared_review_unit_hash": "c9f5e2d5f2aa3c90fcded3f17e12d7f10ef8c924d3878a11e2a4d8750da72cdd", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 14183, - "estimated_input_tokens": 3546, - "estimated_eval_tokens": 7348 + "serialized_chars": 14178, + "estimated_input_tokens": 3545, + "estimated_eval_tokens": 7346 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 6, - "serialized_chars": 14183, - "estimated_input_tokens": 3546, - "estimated_eval_tokens": 7348 + "serialized_chars": 14178, + "estimated_input_tokens": 3545, + "estimated_eval_tokens": 7346 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are four separate model-addition PRs that only overlap on common Transformers scaffolding (docs, auto registries, test/repo plumbing). They do not look like duplicates of one another.", + "summary": "These PRs all add different model integrations (ViT NEPA, Molmo, Molmo2, CTSM) and only share broad scaffolding like docs/auto-registry updates, so they are not duplicates.", "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43451 (Molmo2) is the most developed representative of the cluster, with the highest review activity and the broadest implementation footprint, but it is still a distinct model-addition PR rather than a duplicate of the others.", + "canonical_pr_reason": "PR 43451 (Molmo2) is the most substantial and active of the set, so it is the best anchor if a single representative PR is needed, but it is not the same change as the others.", "best_issue_reason": null, - "best_pr_reason": "PR 43451 is the strongest cluster representative: it has the largest review footprint and a full model/docs/tests integration, making it the best single PR to anchor the cluster even though the others are unrelated additions.", + "best_pr_reason": "PR 43451 is the strongest standalone PR in the cluster: largest review activity and a full model stack addition, making it the best representative despite the cluster containing unrelated model additions.", "soft_edge_verdicts": [ { "left": "pull_request:42978", "right": "pull_request:43448", "accept": false, - "reason": "Different model families (ViT NEPA vs Molmo). Shared files are generic model-registration and docs scaffolding, not the same bug or change." + "reason": "Different model additions (ViT NEPA vs Molmo) with distinct model files and tests; shared registry/docs edits are incidental." }, { "left": "pull_request:42978", "right": "pull_request:43451", "accept": false, - "reason": "ViT NEPA and Molmo2 are distinct model integrations with different codepaths and filenames; overlap is only in shared auto/docs plumbing." + "reason": "ViT NEPA and Molmo2 are separate architectures and code paths; overlap is only in generic auto/docs plumbing." }, { "left": "pull_request:42978", "right": "pull_request:45490", "accept": false, - "reason": "ViT NEPA is a vision model add-on, while CTSM is a timeseries model. They share only broad repository boilerplate, so they are not mergeable as one PR." + "reason": "ViT NEPA and CTSM are unrelated model integrations, not one underlying bug or change." }, { "left": "pull_request:43448", "right": "pull_request:43451", "accept": false, - "reason": "Molmo and Molmo2 are related by name but are separate model releases with different implementation files, processors, and tests; not the same concrete change." + "reason": "Molmo and Molmo2 are different models with different implementations; they cannot plausibly be merged as one fix/change." }, { "left": "pull_request:43448", "right": "pull_request:45490", "accept": false, - "reason": "Molmo and CTSM target unrelated domains and codepaths. The shared files are generic registry/docs changes, which is insufficient for duplicate acceptance." + "reason": "Molmo and CTSM touch separate model families and code paths; common files are only shared integration points." }, { "left": "pull_request:43451", "right": "pull_request:45490", "accept": false, - "reason": "Molmo2 and CTSM are different model additions in different subsystems (multimodal vs timeseries); they do not fix the same underlying problem and would not plausibly be merged into one PR." + "reason": "Molmo2 and CTSM are distinct model additions with different affected modules and tests, so this is not a duplicate pair." } ] }, @@ -18015,58 +18081,58 @@ "pull_request:43665|pull_request:46102", "pull_request:43665|pull_request:44815" ], - "prepared_review_unit_hash": "3bd095dbc49925ec46064b6fe12bc2771d6ee887223847392eca7fe0ecd20514", + "prepared_review_unit_hash": "8ca51f6891c4288b5e18dbfb41ab24e597ff2308c60248069ba629d146fbacf9", "component_budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 9387, - "estimated_input_tokens": 2347, - "estimated_eval_tokens": 4950 + "serialized_chars": 9391, + "estimated_input_tokens": 2348, + "estimated_eval_tokens": 4952 }, "budget": { "node_count": 4, "item_count": 4, "soft_pair_count": 4, - "serialized_chars": 9387, - "estimated_input_tokens": 2347, - "estimated_eval_tokens": 4950 + "serialized_chars": 9391, + "estimated_input_tokens": 2348, + "estimated_eval_tokens": 4952 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The cluster is mostly superficial similarity from generic fix titles; the PRs touch unrelated areas (Whisper tokenization/tests, CLIP/ViT tests, FP8 dequant/loading, and FSDP tied weights). No pair looks like the same underlying change.", - "confidence": 0.95, + "summary": "These PRs are not duplicates: they touch different subsystems and fix unrelated problems. The only clearly substantive PR among them is the FP8 dequantization fix.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "No single PR is a true canonical duplicate target here because the items are unrelated; if forced to pick the most representative, PR #44815 is the most concrete, code-path-focused fix with a clear implementation and test surface.", + "canonical_pr_reason": "No single canonical duplicate: the PRs are unrelated (Whisper decoding/tests, CLIP/ViT test stubs, FP8 dequantization, and FSDP tied-weights cleanup/fix).", "best_issue_reason": null, - "best_pr_reason": "PR #44815 is the strongest standalone change: it addresses a specific FP8/dequantization path, spans the relevant implementation files, and is more substantive than the test-only or generic-fix PRs.", + "best_pr_reason": "PR 44815 is the strongest representative because it contains a concrete multi-file code fix in model loading and FP8 quantization paths, with supporting tests.", "soft_edge_verdicts": [ { "left": "pull_request:43291", "right": "pull_request:43665", "accept": false, - "reason": "Different subsystems and bug classes: Whisper tokenizer decode logic vs. CLIP/ViT test edits. The shared 'fix' framing is only superficial." + "reason": "Different areas and different changes: Whisper tokenizer logic vs unrelated CLIP/ViT test edits." }, { "left": "pull_request:43291", "right": "pull_request:44815", "accept": false, - "reason": "Whisper decoding/timestamp handling is unrelated to FP8 dequantization and model loading changes." + "reason": "Whisper decoding/timestamp handling is unrelated to FP8 dequantization and model-loading changes." }, { "left": "pull_request:43665", "right": "pull_request:46102", "accept": false, - "reason": "Generic test changes in CLIP/ViT do not match FSDP tied-weights/runtime behavior changes." + "reason": "Placeholder test failures in CLIP/ViT do not match an FSDP tied-weights fix." }, { "left": "pull_request:43665", "right": "pull_request:44815", "accept": false, - "reason": "Test-only CLIP/ViT edits are not the same underlying change as FP8 dequant/loading fixes." + "reason": "Generic test changes are not the same underlying bug as the FP8 dequantization/model-loading fix." } ] }, @@ -18103,70 +18169,70 @@ "pull_request:45932|pull_request:46100", "pull_request:46100|pull_request:46114" ], - "prepared_review_unit_hash": "dfc43ec2cc1e156d0ced3b81245bc88cd346575ac06f915f73adb125f574287b", + "prepared_review_unit_hash": "81087537cf084555e9063b262cb4072fe42a4975497e8f528d533716c384b6d5", "component_budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 5, - "serialized_chars": 11343, - "estimated_input_tokens": 2836, - "estimated_eval_tokens": 5928 + "serialized_chars": 11339, + "estimated_input_tokens": 2835, + "estimated_eval_tokens": 5926 }, "budget": { "node_count": 5, "item_count": 5, "soft_pair_count": 5, - "serialized_chars": 11343, - "estimated_input_tokens": 2836, - "estimated_eval_tokens": 5928 + "serialized_chars": 11339, + "estimated_input_tokens": 2835, + "estimated_eval_tokens": 5926 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster is mostly a mix of unrelated placeholder/test PRs. The only true duplicate pair is the ViT test-migration PRs 46100 and 46114, which have the same patch; the other PRs target different workflows or test/setup changes and should stay separate.", - "confidence": 0.94, + "summary": "One clear PR duplicate pair exists: 46100 and 46114 are the same ViT test-migration edit. The other PRs are lookalikes only and differ in scope/content, so they should not be merged as duplicates.", + "confidence": 0.96, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46100 is the earlier of the exact duplicate ViT test-migration PRs and is the cleanest anchor for that pair.", + "canonical_pr_reason": "PR 46100 is the cleanest representative of the duplicated ViT test-migration change: it matches 46114 line-for-line in tests/models/vit/test_modeling_vit.py, while the rest of the cluster is unrelated.", "best_issue_reason": null, - "best_pr_reason": "PR 46100 is the best representative because it matches 46114 byte-for-byte in the shown diff and is the earliest instance of that change.", + "best_pr_reason": "PR 46100 is the best canonical PR because it is the earlier, straightforward version of the duplicated patch and its diff is identical to 46114.", "soft_edge_verdicts": [ { "left": "pull_request:43488", - "right": "pull_request:46114", + "right": "pull_request:43532", "accept": false, - "reason": "Different changes: 43488 tweaks setup/versioning and repo-bot test plumbing, while 46114 is only a ViT test-migration stub." + "reason": "Same broad repo area and one shared test filename, but the changes are unrelated: 43488 is a setup/dependency/version tweak, while 43532 changes workflow/test infrastructure across several files." }, { "left": "pull_request:43488", - "right": "pull_request:43532", + "right": "pull_request:46114", "accept": false, - "reason": "Both are dummy/internal PRs, but the patches are unrelated: setup/versioning vs workflow/comment-ci changes." + "reason": "Different underlying changes. 43488 is a repo-bot/setup PR with dependency/version edits; 46114 is a tiny ViT test-migration change in tests/models/vit/test_modeling_vit.py." }, { "left": "pull_request:43532", "right": "pull_request:46114", "accept": false, - "reason": "Unrelated code paths: self-comment CI/workflow plumbing versus a ViT test file stub." + "reason": "They both touch test-related files, but the concrete change is different: 43532 modifies workflows and test utilities, while 46114 only adds failing ViT test stubs." }, { "left": "pull_request:45932", "right": "pull_request:46100", "accept": false, - "reason": "Not the same underlying change: 45932 spans multiple model/config/workflow files, while 46100 only adds failing assertions to one ViT test file." + "reason": "Not the same bug/change. 45932 is a broad multi-model/config/workflow PR, whereas 46100 only edits a single ViT test file with dummy assertions." }, { "left": "pull_request:46100", "right": "pull_request:46114", "accept": true, - "reason": "Same underlying change: both patches modify the same ViT test file in the same way and appear to be duplicate test-migration PRs." + "reason": "These are the same underlying change: identical edits to tests/models/vit/test_modeling_vit.py adding the same test-migration assertions, so they could plausibly be merged into one PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded overall. The packet clearly supports 46100 and 46114 as exact duplicate ViT test-migration PRs, with 46100 created earlier. The other pairwise rejections are conservative and the listed changes do not look mergeable into one PR for the same bugfix." + "feedback": "The cluster summary is grounded in the packet. PRs 46100 and 46114 have identical diffs in tests/models/vit/test_modeling_vit.py, so treating them as the duplicate pair is well supported. The other pairs are rejected conservatively based on differing scope and content, not just filename overlap." }, "evaluator_used": true, "retried": false, @@ -18193,12 +18259,12 @@ "soft_pairs": [ "pull_request:43612|pull_request:43613" ], - "prepared_review_unit_hash": "8494015f0f9b45680bfeb0b449b07d19b05165a4f99b7ea9f06584d563e16609", + "prepared_review_unit_hash": "ec6d1d9aa72adba06ef7bc037729bb4f9ba9e9783de8cb9341faf5cd2f9360a6", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6082, + "serialized_chars": 6083, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -18206,7 +18272,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 6082, + "serialized_chars": 6083, "estimated_input_tokens": 1521, "estimated_eval_tokens": 3298 }, @@ -18215,18 +18281,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are related feature additions in the same area, but they implement different pipelines for different segmentation tasks and model sets, so they should not be treated as duplicates.", + "summary": "These PRs share pipeline/documentation scaffolding but implement different features: promptable concept segmentation vs promptable visual segmentation. They are related, not duplicates.", "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43613 is the broader anchor if one must pick a representative, but it is still a separate feature from 43612 rather than the same change.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 43613 is the stronger standalone representative because it covers a broader visual-segmentation integration and has the larger implementation footprint, but it is not a duplicate of 43612.", + "best_pr_reason": null, "soft_edge_verdicts": [ { "left": "pull_request:43612", "right": "pull_request:43613", "accept": false, - "reason": "Different feature PRs: promptable concept segmentation vs promptable visual segmentation, with distinct pipeline files, model integrations, and tests; they are related but not the same concrete change." + "reason": "Both add similar segmentation-pipeline plumbing, but they target different underlying changes and code paths: concept segmentation vs visual segmentation, with different model integrations and test files. They should stay separate." } ] }, @@ -18256,7 +18322,7 @@ "soft_pairs": [ "pull_request:43651|pull_request:43663" ], - "prepared_review_unit_hash": "ee703b137d362bb95fb0c4dfca0e15d85d8dc56d57516dbd6f660ff72776372f", + "prepared_review_unit_hash": "47e97af7371d620064507c8470b80e5ab10877bd2efa10299f797e3dfe376c23", "component_budget": { "node_count": 2, "item_count": 2, @@ -18278,18 +18344,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are two separate trainer PRs in the same file but on different code paths: one changes gradient-accumulation loss scaling, the other extracts signature-column logic for model input filtering. They are not the same underlying change.", - "confidence": 0.96, + "summary": "Both PRs modify `src/transformers/trainer.py` and add override hooks, but they target unrelated trainer behaviors: gradient-accumulation loss scaling vs. signature-based column filtering.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR for a duplicate cluster: 43651 and 43663 address unrelated trainer behaviors and should not be merged as one change.", + "canonical_pr_reason": "No single canonical PR: these are separate trainer extension-point changes, not duplicate fixes for the same code path.", "best_issue_reason": null, - "best_pr_reason": "43651 is the stronger representative PR if one must be chosen, because it makes a concrete behavior change in training_step rather than an internal helper extraction.", + "best_pr_reason": "PR 43651 is the more self-contained change, but it is still distinct from 43663 and does not subsume it.", "soft_edge_verdicts": [ { "left": "pull_request:43651", "right": "pull_request:43663", "accept": false, - "reason": "Different functionality: 43651 adjusts loss scaling during gradient accumulation, while 43663 refactors signature-column inspection for filtering inputs. Same file is not enough to treat them as the same change." + "reason": "Reject: same file and similar helper-style refactor, but the underlying behavior is different. 43651 changes loss scaling during training steps; 43663 changes which input columns are kept by inspecting the model signature." } ] }, @@ -18319,12 +18385,12 @@ "soft_pairs": [ "pull_request:43656|pull_request:43842" ], - "prepared_review_unit_hash": "54a80682d5c32b395bd5d3ba26ad64a06bc51216280731bdbd2cfc6600b82575", + "prepared_review_unit_hash": "127355f5fb6fe5ec7cef7191486be3a58e65d6bbe3849c33276fec28a4aa93d0", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3663, + "serialized_chars": 3664, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -18332,7 +18398,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 3663, + "serialized_chars": 3664, "estimated_input_tokens": 916, "estimated_eval_tokens": 2088 }, @@ -18341,28 +18407,25 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are very likely variants of the same CLI import-time TypeAdapter/type-annotation NameError fix in src/transformers/cli/serve.py.", - "confidence": 0.86, + "summary": "Both PRs touch `src/transformers/cli/serve.py` and mention a TypeAdapter-related NameError, but they appear to fix different annotation/import problems rather than the same concrete change.", + "confidence": 0.8, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 43842 is the cleaner, more targeted fix: it explicitly keeps TypeAdapter behind TYPE_CHECKING and quotes the annotation to avoid runtime import errors when pydantic is absent.", + "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "PR 43842 is the best representative of the cluster because it directly addresses the concrete failure mode described in its title and diff, with minimal unrelated typing cleanup.", + "best_pr_reason": "PR #43842 is the more direct, narrowly scoped fix for the explicit `TypeAdapter` NameError when `pydantic` is absent; PR #43656 is a broader typing/annotation cleanup and is not the same patch.", "soft_edge_verdicts": [ { "left": "pull_request:43656", "right": "pull_request:43842", - "accept": true, - "reason": "Both PRs modify the same file to prevent the same TypeAdapter-related import/annotation NameError in the CLI; they look like two implementations of the same underlying bug fix and could plausibly be merged into one PR." + "accept": false, + "reason": "Same file and related symptom, but different fixes: #43656 is a broader annotations/__future__ typing change, while #43842 specifically gates `TypeAdapter` behind `TYPE_CHECKING` and quotes the annotation. They are not the same concrete code-path change." } ] }, - "evaluator_result": { - "accept": true, - "feedback": "The summary is grounded in the packet: both PRs touch `src/transformers/cli/serve.py` and both are titled around a `TypeAdapter`/annotation NameError in the CLI. The mergeability claim is conservative enough because the diffs appear compatible and address the same runtime-import hazard, even though PR 43656 also includes broader typing cleanup." - }, - "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 }, @@ -18548,400 +18611,364 @@ "pull_request:44044|pull_request:44154", "pull_request:44024|pull_request:44076" ], - "prepared_review_unit_hash": "7b4a39cbfbbc57e72be9a8e1252e80d7fa6937b53013aee703092cc7198efb67", + "prepared_review_unit_hash": "6e73c3f14aebf60382e5b31b10915d870b5e7187e677ad82c26dd1da0bd5918b", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84504, - "estimated_input_tokens": 21126, - "estimated_eval_tokens": 42508 + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 }, "budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 72, - "serialized_chars": 76935, - "estimated_input_tokens": 19234, - "estimated_eval_tokens": 38724 + "serialized_chars": 76942, + "estimated_input_tokens": 19236, + "estimated_eval_tokens": 38728 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "This cluster is mostly a topic group around converting model-specific forward paths to standardized output tracing decorators. A few PRs are true duplicates within the same model family (notably GPT-J/CodeGen and GPT-Neo), but most only share the umbrella issue #43979 and are not duplicates.", - "confidence": 0.88, - "canonical_issue_reason": "Issue #43979 is the shared umbrella for the standardized output-tracing migration across these models.", - "canonical_pr_reason": "PR #44722 is the strongest representative of the GPT-J/CodeGen duplicate subgroup: it targets the common umbrella issue, touches the same concrete files as the other GPT-J variants, and appears to be the latest consolidated version of that exact change.", - "best_issue_reason": "#43979 is the best global issue anchor because all of these PRs are variations of that output-tracing migration program.", - "best_pr_reason": "#44722 is the best global PR anchor among the actual duplicates, but only for the GPT-J/CodeGen subgroup; the rest of the cluster is too heterogeneous to collapse into one PR.", + "summary": "This cluster is mostly a batch of model-specific output-tracing refactors, not one single duplicate set. The true duplicate subclusters are GPT-Neo (44018/44068), GPT-J/CodeGen (44066/44085/44722), and the Flaubert-only portion of 44101/44116; most other pairs are just the same refactor theme across different models.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "44066 is the clearest anchor for the main duplicate subcluster: it directly targets #43979, touches the same GPT-J/CodeGen files as 44722/44085, and its diff matches the standardized output-tracing refactor closely.", + "best_issue_reason": null, + "best_pr_reason": "44066 is the best representative PR in the cluster because it has direct issue linkage, exact file overlap with the closest duplicates, and the most straightforward metadata for the GPT-J/CodeGen tracing change.", "soft_edge_verdicts": [ { "left": "pull_request:43996", "right": "pull_request:44085", "accept": false, - "reason": "Both are output-tracing refactors, but for different models/files; same umbrella issue is not enough to make them duplicates." + "reason": "Different models and code paths; only the generic output-tracing theme overlaps." }, { "left": "pull_request:43996", "right": "pull_request:44044", "accept": false, - "reason": "Different model families and code paths; only the refactor theme matches." + "reason": "CVT/FNet vs DeBERTa are unrelated concrete changes despite similar refactor style." }, { "left": "pull_request:44066", "right": "pull_request:44085", "accept": true, - "reason": "Same GPT-J/CodeGen output-tracing refactor; same concrete code paths and mergeable as one PR." + "reason": "Same GPT-J output-tracing refactor in the same file, with the same issue target and nearly identical changes." }, { "left": "pull_request:44007", "right": "pull_request:44072", "accept": false, - "reason": "Both standardize output tracing, but they affect different vision model implementations." + "reason": "ResNet/RegNet image backbones are separate code paths, so this is only thematic similarity." }, { "left": "pull_request:44072", "right": "pull_request:44722", "accept": false, - "reason": "Different models and files; this is only a shared refactor pattern." + "reason": "EfficientNet and GPT-J are unrelated models; no shared concrete bug/change." }, { "left": "pull_request:44066", "right": "pull_request:44072", "accept": false, - "reason": "Different model families; no shared concrete bug/change." + "reason": "CodeGen/GPT-J tracing vs EfficientNet tracing are different model implementations." }, { "left": "pull_request:44013", "right": "pull_request:44044", "accept": false, - "reason": "Different architectures; shared issue target alone does not make them duplicates." + "reason": "MobileNetV2 and DeBERTaV2 are distinct changes with no shared code path." }, { "left": "pull_request:44066", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J/CodeGen versus MGP-STR; same theme, different underlying change." + "reason": "GPT-J/CodeGen tracing and MGP-STR tracing are separate model-specific refactors." }, { "left": "pull_request:44066", "right": "pull_request:44071", "accept": false, - "reason": "Different model files and code paths; not the same fix." + "reason": "GPT-J/CodeGen and MPT touch different model stacks and different output plumbing." }, { "left": "pull_request:44018", "right": "pull_request:44068", "accept": true, - "reason": "Same GPT-Neo output-tracing refactor for the same model file and path." + "reason": "Both are GPT-Neo output-tracing migrations in the same file and appear to be the same concrete refactor." }, { "left": "pull_request:44066", "right": "pull_request:44068", "accept": false, - "reason": "GPT-J/CodeGen and GPT-Neo are separate fixes, even if both are tracing refactors." + "reason": "GPT-J/CodeGen and GPT-Neo are different model files and different changes." }, { "left": "pull_request:44018", "right": "pull_request:44066", "accept": false, - "reason": "Different model families; shared umbrella issue is insufficient." + "reason": "GPT-Neo vs GPT-J/CodeGen are separate model implementations." }, { "left": "pull_request:44019", "right": "pull_request:44722", "accept": false, - "reason": "ResNet output-tracing refactor versus GPT-J/CodeGen refactor; unrelated concrete code paths." + "reason": "ResNet and GPT-J are unrelated model changes." }, { "left": "pull_request:44086", "right": "pull_request:44722", "accept": false, - "reason": "MGP-STR versus GPT-J/CodeGen; same general pattern but not the same change." + "reason": "MGP-STR and GPT-J/CodeGen are different concrete code paths." }, { "left": "pull_request:44019", "right": "pull_request:44085", "accept": false, - "reason": "Different models and implementations; no duplicate evidence beyond theme." + "reason": "ResNet and GPT-J are unrelated despite the shared output-tracing theme." }, { "left": "pull_request:44018", "right": "pull_request:44019", "accept": false, - "reason": "GPT-Neo and ResNet are separate refactors in different code paths." + "reason": "GPT-Neo and ResNet are separate model-specific refactors." }, { "left": "pull_request:44019", "right": "pull_request:44071", "accept": false, - "reason": "ResNet versus MPT; only the output-tracing initiative matches." + "reason": "ResNet and MPT do not share the same underlying bug/change." }, { "left": "pull_request:44068", "right": "pull_request:44722", "accept": false, - "reason": "GPT-Neo refactor versus GPT-J/CodeGen refactor; not the same underlying fix." + "reason": "GPT-Neo and GPT-J/CodeGen are different model families and different PRs." }, { "left": "pull_request:44068", "right": "pull_request:44116", "accept": false, - "reason": "Different model families and files; no duplicate relationship." + "reason": "GPT-Neo and Flaubert are unrelated code paths." }, { "left": "pull_request:44071", - "right": "pull_request:44722", + "right": "pull_request:44086", "accept": false, - "reason": "MPT and GPT-J/CodeGen are separate model-specific changes." + "reason": "MPT and MGP-STR are different model implementations." }, { "left": "pull_request:44018", "right": "pull_request:44116", "accept": false, - "reason": "Different models; only the refactor style is shared." + "reason": "GPT-Neo and Flaubert are unrelated model refactors." }, { "left": "pull_request:44018", "right": "pull_request:44086", "accept": false, - "reason": "GPT-Neo and MGP-STR are unrelated concrete changes." + "reason": "GPT-Neo and MGP-STR are separate concrete changes." }, { "left": "pull_request:44018", "right": "pull_request:44085", "accept": false, - "reason": "GPT-Neo versus GPT-J/CodeGen; not the same bug/change." + "reason": "GPT-Neo and GPT-J are different model code paths." }, { "left": "pull_request:44068", "right": "pull_request:44086", "accept": false, - "reason": "Separate model refactors with no shared code path." + "reason": "GPT-Neo and MGP-STR are unrelated." }, { "left": "pull_request:44071", "right": "pull_request:44085", "accept": false, - "reason": "MPT and GPT-J/CodeGen are not duplicates." + "reason": "MPT and GPT-J are different model families." }, { "left": "pull_request:44059", "right": "pull_request:44086", "accept": false, - "reason": "GPT-2 and MGP-STR are different model implementations." + "reason": "GPT2 and MGP-STR are separate model-specific changes." }, { "left": "pull_request:44085", "right": "pull_request:44086", "accept": false, - "reason": "GPT-J/CodeGen versus MGP-STR; same theme only." + "reason": "GPT-J and MGP-STR are different code paths." }, { "left": "pull_request:44098", "right": "pull_request:44101", "accept": false, - "reason": "ViLT and Flaubert/XLM are unrelated model-specific refactors." + "reason": "ViLT and Flaubert/XLM are different model implementations." }, { "left": "pull_request:44059", "right": "pull_request:44722", "accept": false, - "reason": "GPT-2 and GPT-J/CodeGen differ in concrete code path and change set." + "reason": "GPT2 and GPT-J/CodeGen are not the same underlying change." }, { - "left": "pull_request:44018", + "left": "pull_request:44068", "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo and MPT are distinct model refactors." - }, - { - "left": "pull_request:44017", - "right": "pull_request:44722", - "accept": false, - "reason": "Segformer and GPT-J/CodeGen are different underlying changes." + "reason": "GPT-Neo and MPT are different models and different output handling." }, { - "left": "pull_request:44017", + "left": "pull_request:44073", "right": "pull_request:44161", "accept": false, - "reason": "Segformer versus LongT5; only the broad migration theme overlaps." + "reason": "VisualBert and LongT5 are unrelated concrete changes." }, { - "left": "pull_request:44017", - "right": "pull_request:44073", + "left": "pull_request:44059", + "right": "pull_request:44068", "accept": false, - "reason": "Segformer and VisualBert are separate model-specific changes." + "reason": "GPT2 and GPT-Neo are separate model families." }, { - "left": "pull_request:44068", + "left": "pull_request:44059", "right": "pull_request:44071", "accept": false, - "reason": "GPT-Neo versus MPT; no duplicate evidence." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44073", - "accept": false, - "reason": "RegNet and VisualBert are unrelated concrete refactors." - }, - { - "left": "pull_request:44071", - "right": "pull_request:44076", - "accept": false, - "reason": "MPT and ImageGPT touch different model implementations." - }, - { - "left": "pull_request:44074", - "right": "pull_request:44101", - "accept": false, - "reason": "TextNet and Flaubert/XLM are separate model-specific changes." + "reason": "GPT2 and MPT are different model code paths." }, { - "left": "pull_request:44056", + "left": "pull_request:44101", "right": "pull_request:44161", "accept": false, - "reason": "MPNet and LongT5 are different code paths; not the same PR." + "reason": "Flaubert/XLM and LongT5 are unrelated changes." }, { "left": "pull_request:43997", "right": "pull_request:44017", "accept": false, - "reason": "RegNet and Segformer are distinct refactors." + "reason": "RegNet and Segformer are different backbone implementations." }, { "left": "pull_request:43997", - "right": "pull_request:44056", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 do not share the same bug/change." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", "accept": false, - "reason": "RegNet and MPNet are unrelated concrete fixes." + "reason": "TextNet and Flaubert are separate model refactors." }, { "left": "pull_request:44071", "right": "pull_request:44129", "accept": false, - "reason": "MPT and SpeechT5 are separate model changes." + "reason": "MPT and SpeechT5 are different model families." }, { "left": "pull_request:44010", "right": "pull_request:44071", "accept": false, - "reason": "SqueezeBERT and MPT do not share the same underlying bug/change." + "reason": "SqueezeBERT and MPT are unrelated concrete changes." }, { "left": "pull_request:44010", "right": "pull_request:44073", "accept": false, - "reason": "Different models and implementations; only the output-tracing theme matches." + "reason": "SqueezeBERT and VisualBert are different code paths." }, { "left": "pull_request:44018", "right": "pull_request:44129", "accept": false, - "reason": "GPT-Neo and SpeechT5 are separate code paths." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44074", - "accept": false, - "reason": "RegNet and TextNet are unrelated model-specific refactors." - }, - { - "left": "pull_request:43997", - "right": "pull_request:44129", - "accept": false, - "reason": "RegNet and SpeechT5 are not duplicates." - }, - { - "left": "pull_request:44071", - "right": "pull_request:44129", - "accept": false, - "reason": "MPT and SpeechT5 affect different models and behaviors." - }, - { - "left": "pull_request:44010", - "right": "pull_request:44018", - "accept": false, - "reason": "SqueezeBERT and GPT-Neo are separate changes." + "reason": "GPT-Neo and SpeechT5 are unrelated model changes." }, { "left": "pull_request:43996", "right": "pull_request:44161", "accept": false, - "reason": "CVT/FNet and LongT5 are different model implementations." + "reason": "CVT/FNet and LongT5 are unrelated models." }, { "left": "pull_request:43996", "right": "pull_request:44017", "accept": false, - "reason": "CVT/FNet versus Segformer; same migration theme, different fixes." + "reason": "CVT/FNet and Segformer are separate implementations." }, { "left": "pull_request:44010", "right": "pull_request:44044", "accept": false, - "reason": "SqueezeBERT and DeBERTa-v2 are unrelated concrete refactors." + "reason": "SqueezeBERT and DeBERTaV2 are different model families." }, { "left": "pull_request:44044", "right": "pull_request:44074", "accept": false, - "reason": "DeBERTa-v2 and TextNet are different model-specific changes." + "reason": "DeBERTaV2 and TextNet are unrelated changes." }, { "left": "pull_request:44066", "right": "pull_request:44722", "accept": true, - "reason": "Same GPT-J/CodeGen output-tracing refactor; same code paths and effectively the same change." + "reason": "Same GPT-J/CodeGen standardized output-tracing change with the same files and issue target." }, { "left": "pull_request:44085", "right": "pull_request:44722", "accept": true, - "reason": "Same GPT-J output-tracing refactor; the title mismatch does not change the underlying file-level change." + "reason": "Both are the same GPT-J output-tracing refactor in the same file; one is just a narrower variant." }, { "left": "pull_request:44085", "right": "pull_request:44129", "accept": false, - "reason": "GPT-J and SpeechT5 are separate model changes." + "reason": "GPT-J and SpeechT5 are unrelated model-specific refactors." }, { "left": "pull_request:43996", "right": "pull_request:44007", "accept": false, - "reason": "CVT/FNet versus RegNet/ResNet/RT-DETR-ResNet are different fixes." + "reason": "Different image-model code paths; only the general refactor pattern is shared." }, { "left": "pull_request:43996", "right": "pull_request:44129", "accept": false, - "reason": "Different models and code paths; only the umbrella refactor theme overlaps." + "reason": "CVT/FNet and SpeechT5 are unrelated concrete changes." }, { "left": "pull_request:44044", "right": "pull_request:44129", "accept": false, - "reason": "DeBERTa-v2 and SpeechT5 are not the same bug/change." + "reason": "DeBERTaV2 and SpeechT5 do not share the same underlying bug/change." }, { "left": "pull_request:44044", "right": "pull_request:44154", "accept": false, - "reason": "DeBERTa-v2 and VITS are separate model-specific refactors." + "reason": "DeBERTaV2 and ViTS are different model code paths." }, { "left": "pull_request:44024", "right": "pull_request:44076", "accept": false, - "reason": "FocalNet and ImageGPT are unrelated concrete changes; same issue family only." + "reason": "FocalNet and ImageGPT are different models; this is only a shared refactor theme." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44116", + "accept": true, + "reason": "Both touch the same Flaubert attention/output-tracing code path; 44101 just bundles the Flaubert change with XLM." } ] }, "evaluator_result": { "accept": true, - "feedback": "Grounded and appropriately conservative. The summary stays within the packet evidence: it treats #43979 as the umbrella issue, limits duplicate claims to the GPT-J/CodeGen and GPT-Neo subgroups, and rejects cross-model pairs unless there is shared file-level/change-set evidence. I don\u2019t see overstatement or an unmergeable PR-pair claim." + "feedback": "The summary is grounded in the packet and stays conservative about the cluster\u2019s structure. The accepted duplicate subclusters are supported by the diff previews and shared-file/issue-target evidence, and the rejected pairs mostly stay at the level of thematic similarity across different model code paths. I don\u2019t see an overstatement that would warrant rejection." }, "evaluator_used": true, "retried": false, @@ -18976,64 +19003,64 @@ "pull_request:43997|pull_request:44010", "pull_request:44010|pull_request:44129" ], - "prepared_review_unit_hash": "2d14d642ee652ef800c1717fcce56314edb0c00ca2d2ba45b544f5f1844fe158", + "prepared_review_unit_hash": "d07bc2bd0ba655e069bdb1888c2bb61199d8e70a2ab1f16f62e0ae67904e5160", "component_budget": { "node_count": 28, "item_count": 28, "soft_pair_count": 77, - "serialized_chars": 84504, - "estimated_input_tokens": 21126, - "estimated_eval_tokens": 42508 + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 }, "budget": { "node_count": 6, "item_count": 6, "soft_pair_count": 5, - "serialized_chars": 13069, - "estimated_input_tokens": 3268, - "estimated_eval_tokens": 6792 + "serialized_chars": 13068, + "estimated_input_tokens": 3267, + "estimated_eval_tokens": 6790 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": true, "analyst_result": { - "summary": "These PRs are all model-specific refactors to standardized output tracing/capture, but they touch different architectures and code paths, so they are not true duplicates.", - "confidence": 0.97, + "summary": "These PRs are all model-specific refactors to standardized output tracing/capture, but they target different model code paths and should not be merged as duplicates.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44129 is the strongest representative because it has an explicit issue target and a concrete output-tracing refactor in SpeechT5, while the others are similar model-local migrations without evidence they address the same change.", + "canonical_pr_reason": "PR #44129 is the strongest representative because it has an explicit linked issue and the largest, most concrete output-tracing refactor in a single model implementation.", "best_issue_reason": null, - "best_pr_reason": "PR 44129 is the best fit for the cluster representative: it is clearly scoped, has the most explicit tracking context, and shows a substantive but model-specific implementation change.", + "best_pr_reason": "PR #44129 is the best single representative of the cluster, since it is the only one with an explicit issue target and a clearly scoped model-specific implementation change.", "soft_edge_verdicts": [ { "left": "pull_request:44073", "right": "pull_request:44074", "accept": false, - "reason": "Both are output-tracing refactors, but they modify different models (VisualBert vs TextNet) and different implementations; same theme, not the same concrete change." + "reason": "Both are standardized output-capture refactors, but they touch different models (VisualBert vs TextNet) and different forward/output code paths." }, { "left": "pull_request:44056", "right": "pull_request:44073", "accept": false, - "reason": "MPNet and VisualBert have separate model-specific output-capture refactors with different classes and wiring, so they are not mergeable as one fix." + "reason": "Same refactor theme, but MPNet and VisualBert are unrelated model implementations with distinct output-tracing paths." }, { "left": "pull_request:44056", "right": "pull_request:44074", "accept": false, - "reason": "These touch different model code paths (MPNet vs TextNet); they share a refactor pattern but not the same underlying bug or change." + "reason": "These are separate model migrations in MPNet and TextNet; they do not fix the same concrete bug or share a mergeable code path." }, { "left": "pull_request:43997", "right": "pull_request:44010", "accept": false, - "reason": "RegNet and SqueezeBert are unrelated model implementations; both are standardized output-tracing migrations, but they are not duplicates." + "reason": "RegNet and SqueezeBert are different architectures and the PRs change different model-specific forward methods, so they are not duplicates." }, { "left": "pull_request:44010", "right": "pull_request:44129", "accept": false, - "reason": "SqueezeBert and SpeechT5 are different models with different forward-pass logic; the shared output-tracing theme is too broad to treat as the same PR." + "reason": "Both are output-tracing refactors, but they affect different models and different internals; this is broader stylistic similarity, not the same concrete change." } ] }, @@ -19119,286 +19146,304 @@ "pull_request:44002|pull_request:46106", "pull_request:43998|pull_request:46106" ], - "prepared_review_unit_hash": "a40e79ff13c8141585cf4533f3cf58ead8a1ed6924d737abe623b201c799683d", + "prepared_review_unit_hash": "b7400b4d0eed5199b1e9dfda784df12fc4cc1e189f68fcfba4e00c7296b6e36d", "component_budget": { "node_count": 14, "item_count": 14, "soft_pair_count": 45, - "serialized_chars": 43786, - "estimated_input_tokens": 10947, - "estimated_eval_tokens": 22150 + "serialized_chars": 43791, + "estimated_input_tokens": 10948, + "estimated_eval_tokens": 22152 }, "budget": { "node_count": 14, "item_count": 14, "soft_pair_count": 45, - "serialized_chars": 43786, - "estimated_input_tokens": 10947, - "estimated_eval_tokens": 22150 + "serialized_chars": 43791, + "estimated_input_tokens": 10948, + "estimated_eval_tokens": 22152 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These PRs are all in the same broad \u201coutput tracing refactor\u201d theme, but they touch different models and code paths. I found no strong duplicate/merge candidates; the only clearly unrelated item is the docs toctree PR.", - "confidence": 0.73, + "summary": "These items are a set of model-specific output-tracing refactor PRs plus one unrelated docs toctree fix. They share a theme, but not the same concrete bug or mergeable change, so they should not be deduplicated together.", + "confidence": 0.95, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 44025 is the most self-contained representative of the batch: it applies the tracing refactor to a model plus its prompt variant, adds the decorator cleanly, and includes test coverage.", + "canonical_pr_reason": "No single duplicate PR stands out. If one PR must represent the cluster, 44003 is the strongest exemplar because it makes the most substantial output-tracing plumbing changes and spans two related implementations (mamba and falcon_mamba).", "best_issue_reason": null, - "best_pr_reason": "PR 44025 is the best overall representative of this cluster because it is concrete, focused, and more complete than the tiny one-file refactors.", + "best_pr_reason": "44003 is the best representative PR for the cluster\u2019s theme: it is broader than the others, touches the core Mamba path, and includes the capture/output-recorder refactor machinery. It is still not a duplicate of the other PRs.", "soft_edge_verdicts": [ { "left": "pull_request:43998", "right": "pull_request:44002", "accept": false, - "reason": "Different models and code paths (timm_backbone vs upernet); same refactor theme, not the same change." + "reason": "Different model code paths (timm_backbone vs upernet). Same refactor pattern, but not the same underlying change." }, { "left": "pull_request:43998", "right": "pull_request:44000", "accept": false, - "reason": "Different architectures and files; both are output-tracing refactors, but not one underlying bug/change." + "reason": "Different model families and files; both are output-tracing refactors, not one mergeable bug fix." }, { "left": "pull_request:43998", "right": "pull_request:44001", "accept": false, - "reason": "timm_backbone and univnet are unrelated model paths." + "reason": "Unrelated model implementations. Similar helper usage does not make this the same fix." }, { "left": "pull_request:43998", "right": "pull_request:44003", "accept": false, - "reason": "Different model families (timm_backbone vs mamba/falcon_mamba); no shared concrete fix." + "reason": "Different architectures and code paths; this is a shared refactor idea, not the same concrete change." }, { "left": "pull_request:43998", "right": "pull_request:44029", "accept": false, - "reason": "Different model codepaths; same refactor pattern only." + "reason": "Separate model-specific refactors in different code paths; not a duplicate fix." }, { "left": "pull_request:43999", "right": "pull_request:44001", "accept": false, - "reason": "MobileNetV1 and UnivNet are unrelated implementations." + "reason": "MobileNetV1 and UnivNet are unrelated model implementations; same refactor style only." }, { "left": "pull_request:43999", "right": "pull_request:44002", "accept": false, - "reason": "MobileNetV1 vs UperNet; similar output-tracing cleanup, not the same bug." + "reason": "Different models and output types; not the same underlying bug or change." }, { "left": "pull_request:43999", "right": "pull_request:44025", "accept": false, - "reason": "DepthAnything and MobileNetV1 are separate model families with distinct forward paths." + "reason": "Both refactor tracing, but for different model families and files; not mergeable as one bug fix." }, { "left": "pull_request:43999", "right": "pull_request:44026", "accept": false, - "reason": "Different seq2seq wrapper/model codepaths; not mergeable as one fix." + "reason": "Different code paths in different model wrappers; similar plumbing, not duplicate content." }, { "left": "pull_request:43999", "right": "pull_request:44027", "accept": false, - "reason": "MobileNetV1 and SpeechEncoderDecoder are unrelated; same refactor style only." + "reason": "MobileNetV1 and SpeechEncoderDecoder are unrelated; shared tracing refactor theme is too broad." }, { "left": "pull_request:44000", "right": "pull_request:44001", "accept": false, - "reason": "Vision-text dual encoder vs UnivNet: different model domains and outputs." + "reason": "Vision-text dual encoder vs UnivNet are different model classes; not the same fix." }, { "left": "pull_request:44000", "right": "pull_request:44002", "accept": false, - "reason": "Different model codepaths; no shared concrete defect." + "reason": "Different model implementations and semantics; only the refactor style matches." }, { "left": "pull_request:44000", "right": "pull_request:44003", "accept": false, - "reason": "Vision-text dual encoder and mamba/falcon_mamba are unrelated." + "reason": "Separate model-specific output-tracing changes; not one concrete code-path issue." }, { "left": "pull_request:44000", - "right": "pull_request:44004", + "right": "pull_request:44025", "accept": false, - "reason": "CodeGen cache/output-signature change is a different fix from vision-text tracing refactor." + "reason": "Different model families and output contracts; not a duplicate." }, { "left": "pull_request:44001", "right": "pull_request:44002", "accept": false, - "reason": "UnivNet and UperNet are different architectures and forward paths." + "reason": "UnivNet and UperNet are unrelated; this is not the same underlying bug/change." }, { "left": "pull_request:44001", "right": "pull_request:44003", "accept": false, - "reason": "Different model families; only the refactor motif is shared." + "reason": "Different model code paths and outputs; too broad to treat as one duplicate fix." }, { "left": "pull_request:44001", "right": "pull_request:44025", "accept": false, - "reason": "UnivNet vs DepthAnything: separate model-specific tracing changes." + "reason": "DepthAnything and UnivNet are separate model refactors, not a single mergeable PR." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet vs CodeGen are different architectures; shared output-tracing cleanup is not enough." }, { "left": "pull_request:44002", "right": "pull_request:44029", "accept": false, - "reason": "UperNet and RWKV are unrelated codepaths." + "reason": "UperNet and RWKV are unrelated model implementations; not the same underlying issue." }, { "left": "pull_request:44002", "right": "pull_request:44003", "accept": false, - "reason": "UperNet vs mamba/falcon_mamba; no single underlying bug." + "reason": "Different model families and files; similar refactor pattern only." }, { "left": "pull_request:44003", "right": "pull_request:44028", "accept": false, - "reason": "Mamba/FalconMamba and SuperPoint are entirely different model stacks." + "reason": "Mamba/FalconMamba vs SuperPoint are unrelated code paths; not a duplicate change." }, { "left": "pull_request:44003", "right": "pull_request:44004", "accept": false, - "reason": "CodeGen\u2019s forward/cache refactor is not the same change as mamba tracing cleanup." + "reason": "Mamba and CodeGen are different model implementations; same broad goal, not same fix." }, { "left": "pull_request:43998", "right": "pull_request:44004", "accept": false, - "reason": "TimmBackbone and CodeGen are unrelated subsystems." + "reason": "TimmBackbone and CodeGen are unrelated; they only share a refactor motif." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model code paths and output semantics; not mergeable as one PR." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are unrelated model changes, not the same bug." }, { "left": "pull_request:44025", "right": "pull_request:44026", "accept": false, - "reason": "DepthAnything and VisionEncoderDecoder are different models; similar decorator usage is not enough." + "reason": "DepthAnything vs VisionEncoderDecoder are different model wrappers; similar tracing refactor only." }, { "left": "pull_request:44025", "right": "pull_request:44027", "accept": false, - "reason": "DepthAnything and SpeechEncoderDecoder are separate codepaths." + "reason": "Different model families and code paths; not the same concrete fix." }, { "left": "pull_request:44025", "right": "pull_request:44028", "accept": false, - "reason": "DepthAnything and SuperPoint are unrelated model implementations." + "reason": "DepthAnything and SuperPoint are unrelated models; no shared bug to merge." }, { "left": "pull_request:44025", "right": "pull_request:44029", "accept": false, - "reason": "DepthAnything and RWKV are distinct architectures with different tracing semantics." + "reason": "Different implementations and output contracts; not a duplicate change." }, { "left": "pull_request:44025", "right": "pull_request:44030", "accept": false, - "reason": "DepthAnything and DPR do not share the same concrete fix." + "reason": "DepthAnything and DPR are different model paths; same theme, not same fix." }, { "left": "pull_request:44026", "right": "pull_request:44027", "accept": false, - "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are different wrapper models; not a duplicate change." + "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are distinct wrappers; not one underlying bug." }, { "left": "pull_request:44026", "right": "pull_request:44028", "accept": false, - "reason": "Different models and forward paths." + "reason": "Different model subsystems; shared tracing cleanup is too broad." }, { "left": "pull_request:44026", "right": "pull_request:44029", "accept": false, - "reason": "Seq2seq wrapper refactor vs RWKV tracing cleanup; not the same bug." + "reason": "Different model code paths and outputs; not a duplicate PR." }, { "left": "pull_request:44026", "right": "pull_request:44030", "accept": false, - "reason": "VisionEncoderDecoder and DPR are unrelated codepaths." + "reason": "VisionEncoderDecoder and DPR are unrelated model implementations." }, { "left": "pull_request:44027", "right": "pull_request:44028", "accept": false, - "reason": "SpeechEncoderDecoder and SuperPoint are unrelated models." + "reason": "SpeechEncoderDecoder and SuperPoint are unrelated; not the same concrete change." }, { "left": "pull_request:44027", "right": "pull_request:44029", "accept": false, - "reason": "Different architectures and output behavior." + "reason": "Different model families and files; only the high-level refactor pattern matches." }, { "left": "pull_request:44027", "right": "pull_request:44030", "accept": false, - "reason": "SpeechEncoderDecoder and DPR are separate implementations." + "reason": "SpeechEncoderDecoder vs DPR are different code paths, so not a duplicate." }, { "left": "pull_request:44028", "right": "pull_request:44029", "accept": false, - "reason": "SuperPoint and RWKV are different models with no shared concrete defect." + "reason": "SuperPoint and RWKV are unrelated implementations; not the same underlying bug." }, { "left": "pull_request:44028", "right": "pull_request:44030", "accept": false, - "reason": "SuperPoint and DPR are unrelated." + "reason": "SuperPoint and DPR are different model paths; no mergeable overlap beyond refactor style." }, { "left": "pull_request:44029", "right": "pull_request:44030", "accept": false, - "reason": "RWKV and DPR are different model paths; not one mergeable fix." + "reason": "RWKV and DPR are different models; the common theme is too generic to count as a duplicate." }, { "left": "pull_request:44027", "right": "pull_request:46106", "accept": false, - "reason": "SpeechEncoderDecoder model code vs docs toctree; unrelated." + "reason": "A docs toctree fix is unrelated to the SpeechEncoderDecoder model refactor." }, { "left": "pull_request:44026", "right": "pull_request:46106", "accept": false, - "reason": "VisionEncoderDecoder model code vs docs toctree; unrelated." + "reason": "Docs navigation change vs model code refactor; clearly different work." }, { "left": "pull_request:44001", "right": "pull_request:46106", "accept": false, - "reason": "UnivNet model code vs docs navigation; unrelated." + "reason": "UnivNet output-tracing refactor and docs toctree fix are unrelated." }, { "left": "pull_request:44002", "right": "pull_request:46106", "accept": false, - "reason": "UperNet model code vs docs navigation; unrelated." + "reason": "UperNet model code and documentation tree changes are not the same change." }, { "left": "pull_request:43998", "right": "pull_request:46106", "accept": false, - "reason": "TimmBackbone model code vs docs navigation; unrelated." + "reason": "TimmBackbone refactor and docs toctree fix are unrelated artifacts." } ] }, @@ -19431,12 +19476,12 @@ "pull_request:44662|pull_request:45075", "pull_request:45075|pull_request:45401" ], - "prepared_review_unit_hash": "9bd863836ab6534769b405adfa0af57d00a5d4edd1129f330793a4f675991588", + "prepared_review_unit_hash": "4abfa6434f8c980164ae676378a498513b114bdddb517f66e16f62c22896f8ea", "component_budget": { "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10912, + "serialized_chars": 10910, "estimated_input_tokens": 2728, "estimated_eval_tokens": 5712 }, @@ -19444,7 +19489,7 @@ "node_count": 3, "item_count": 3, "soft_pair_count": 3, - "serialized_chars": 10912, + "serialized_chars": 10910, "estimated_input_tokens": 2728, "estimated_eval_tokens": 5712 }, @@ -19453,30 +19498,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Three open PRs add separate model integrations (PenguinVL, DeepSeek-OCR-2, and Voxtral TTS). They share only generic new-model scaffolding and documentation/auto-registration files, not the same underlying bug or change, so none should be deduplicated.", - "confidence": 0.98, + "summary": "All three pull requests are distinct model-addition efforts. They share Transformers registration/docs boilerplate, but they target different model families and code paths, so none look like duplicates.", + "confidence": 0.97, "canonical_issue_reason": null, - "canonical_pr_reason": "No canonical PR: each PR implements a different model family, so there is no duplicate to collapse.", + "canonical_pr_reason": "No single canonical PR: PenguinVL, DeepSeek-OCR-2, and Voxtral TTS are separate model integrations with different architecture, files, and implementation scope.", "best_issue_reason": null, - "best_pr_reason": "No best PR for deduplication; all three are distinct model-addition PRs and should remain separate.", + "best_pr_reason": "No PR is a good global duplicate representative here; the overlap is mostly shared auto-mapping/docs scaffolding rather than the same concrete fix or model implementation.", "soft_edge_verdicts": [ { "left": "pull_request:44662", "right": "pull_request:45401", "accept": false, - "reason": "Different models and codepaths: PenguinVL is a vision-language model, while Voxtral TTS is a text-to-speech model. Shared files are broad integration boilerplate only." + "reason": "Different model families and tasks: PenguinVL is a vision-language model, while Voxtral-4B-TTS is a TTS/audio model. Shared auto-registration files are generic boilerplate, not the same underlying change." }, { "left": "pull_request:44662", "right": "pull_request:45075", "accept": false, - "reason": "Different model families: PenguinVL vs DeepSeek-OCR-2. Overlap is limited to common model-registration and docs scaffolding, not the same feature or fix." + "reason": "PenguinVL and DeepSeek-OCR-2 are separate model integrations with different model code, tests, and docs. The overlap is only in common Transformers plumbing, not a shared bug or feature." }, { "left": "pull_request:45075", "right": "pull_request:45401", "accept": false, - "reason": "Distinct integrations for unrelated models (OCR/document vision vs TTS). The overlap is generic auto-mapping and docs setup, which is not enough to treat them as duplicates." + "reason": "DeepSeek-OCR-2 and Voxtral TTS are unrelated additions for different modalities and code paths. They share repository scaffolding, but not the same concrete implementation or fix." } ] }, @@ -19506,12 +19551,12 @@ "soft_pairs": [ "pull_request:44979|pull_request:45363" ], - "prepared_review_unit_hash": "d4aae647312f0c7977d0dac0a8e0072fbd167e240e39993b04400ff509fb6a9c", + "prepared_review_unit_hash": "2c24f8e5c9489ec50d0ea216751d243a05e702288a9ae919d5d3d3beff1f4297", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4406, + "serialized_chars": 4408, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -19519,7 +19564,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4406, + "serialized_chars": 4408, "estimated_input_tokens": 1102, "estimated_eval_tokens": 2460 }, @@ -19528,18 +19573,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two related but distinct PRs in the fusion space: one introduces a generic module-fusion API, the other wires n-to-1 kernel fusion through `KernelConfig` and model loading. They are conceptually adjacent, but not the same concrete change.", - "confidence": 0.88, + "summary": "These PRs are related to module/kernel fusion, but they are not duplicates: #44979 introduces a standalone Module Fusion API, while #45363 wires n-to-1 kernel fusion into `KernelConfig` and model loading. The latter builds on the same theme but targets a different change surface and code path.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45363 is the better cluster representative because it implements a concrete end-to-end fusion workflow (`KernelConfig`, integrations, modeling utilities, example), whereas PR 44979 is a lower-level API scaffold.", + "canonical_pr_reason": "#44979 is the cleaner foundational change: it defines the core `module_fusion.py` API and tests, with a focused scope around module fusion primitives.", "best_issue_reason": null, - "best_pr_reason": "PR 45363 is the strongest standalone PR for this cluster: it is more complete, user-facing, and has a clearly bounded code path that could be reviewed and merged as one feature.", + "best_pr_reason": "#44979 is the best single representative of the cluster because it captures the core abstraction directly, whereas #45363 is a broader integration of a different fusion mechanism.", "soft_edge_verdicts": [ { "left": "pull_request:44979", "right": "pull_request:45363", "accept": false, - "reason": "They are related feature work but not the same underlying change. 44979 adds a generic module-fusion API scaffold; 45363 adds kernel fusion via `KernelConfig` and model/integration plumbing. The code paths and merge scope differ enough that they should not be treated as duplicates." + "reason": "Reject: one PR adds a general module-fusion API, the other integrates n-to-1 kernel fusion via `KernelConfig`; they are related but not the same concrete fix/change and would not be merged as one PR." } ] }, @@ -19569,7 +19614,7 @@ "soft_pairs": [ "pull_request:45086|pull_request:45317" ], - "prepared_review_unit_hash": "578d3c522798886eade8cc5a94984fda2120aec1d71acc3d47405042a769530f", + "prepared_review_unit_hash": "83a4360990b34c18ba1b52cb88c30b0c0297f4abde4049944bcccca5f5e401f8", "component_budget": { "node_count": 2, "item_count": 2, @@ -19591,18 +19636,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests patch the same Mistral pre-tokenizer AttributeError in the same file and target the same issue; the second PR appears to be a fuller version of the first with added tests, not a different fix.", - "confidence": 0.97, + "summary": "Both pull requests fix the same Mistral regex AttributeError in `_patch_mistral_regex` by switching from `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; the second PR is the more complete version because it adds test coverage.", + "confidence": 0.98, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45317 is the better canonical PR because it applies the same code-path fix as 45086 but is more complete, including tests and a fuller discussion history.", + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same code fix as 45086 plus a test file, making it the more complete and reviewable change.", "best_issue_reason": null, - "best_pr_reason": "PR 45317 is the best representative because it contains the same underlying fix plus test coverage, making it the more complete and merge-ready version of the change.", + "best_pr_reason": "PR 45317 is the best representative because it addresses the same concrete bug, keeps the fix localized to the same code path, and adds regression tests.", "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, same source file, and the diff shows the same code change; 45317 just adds tests." + "reason": "Same underlying bug and same fix in the same file; 45317 is essentially 45086 plus tests, so they are mergeable into one PR." } ] }, @@ -19632,7 +19677,7 @@ "soft_pairs": [ "pull_request:45115|pull_request:45569" ], - "prepared_review_unit_hash": "f910d60e700b93759e5f10add8d6828e9a5f156c77bbbe3986a0de6dfa1f4ca5", + "prepared_review_unit_hash": "bc67d9ee3e80991c780168237506e7233fa1a8ed0353e71f6df8bd583c8441a1", "component_budget": { "node_count": 2, "item_count": 2, @@ -19654,18 +19699,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two open NemotronH PRs overlap in files and model code, but they pursue different changes: one is a structural refactor toward GraniteMoeHybrid-compatible inheritance, while the other splits NemotronH into dense/sparse variants and updates docs/tests. They are related, not duplicates.", - "confidence": 0.86, + "summary": "These are related to NemotronH, but they are not the same change. PR #45115 is a narrow conversion/weight-key refactor, while PR #45569 is a much broader model-architecture split introducing NemotronHDense/NemotronHSparse plus docs, tests, and auto-mapping updates.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45569 is the better representative of the cluster because it makes the broader model-architecture change, adds the dense/sparse model split, and includes the user-facing/docs/test updates.", + "canonical_pr_reason": "#45569 is the best canonical PR because it is the broader, more complete NemotronH rework, adding the dense/sparse model packages and associated integration work.", "best_issue_reason": null, - "best_pr_reason": "PR 45569 is the strongest standalone PR in this cluster: it is more complete, broader in scope, and seems to be the main implementation direction rather than a narrow refactor.", + "best_pr_reason": "#45569 is the best representative of the cluster; #45115 is a smaller conversion-mapping refactor and does not cover the larger architecture split.", "soft_edge_verdicts": [ { "left": "pull_request:45115", "right": "pull_request:45569", "accept": false, - "reason": "These PRs touch some of the same NemotronH files, but they are not the same underlying change. 45115 focuses on refactoring NemotronH to inherit GraniteMoeHybrid and adjust conversion mappings, while 45569 introduces dedicated NemotronH dense/sparse models and dispatcher/docs/test updates. They could not plausibly be merged as one single duplicate fix." + "reason": "They share NemotronH files, but the underlying changes differ: #45115 only adjusts conversion/mapping and naming, while #45569 introduces new dense/sparse model variants, docs, tests, and auto-model wiring. They are not the same concrete PR-sized fix." } ] }, @@ -19695,7 +19740,7 @@ "soft_pairs": [ "pull_request:45332|pull_request:45333" ], - "prepared_review_unit_hash": "adfaf2d2e85886b5391814b17288f2d9ba672e8dbeaa5c3bdaa5d8d831bd1d58", + "prepared_review_unit_hash": "c5b9ae5bdcbd043d0d1c1ab211710bf0fca1805cb8e9088eca93dcebbc04353a", "component_budget": { "node_count": 2, "item_count": 2, @@ -19717,18 +19762,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "These are related but distinct feature PRs: one adds heterogeneous configuration support, the other adds heterogeneous modeling/runtime support. They share infrastructure, but they do not appear to fix the same concrete bug or implement the same change end-to-end, so they should not be merged as duplicates.", + "summary": "These PRs are related but not duplicates: one introduces heterogeneous per-layer config support, while the other builds on that groundwork to add heterogeneous model/runtime handling. They share files and concepts, but they target different code paths and could be merged separately.", "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": "No single canonical PR: 45333 is configuration-only groundwork, while 45332 is the broader modeling/caching feature built on top of it.", + "canonical_pr_reason": "PR 45332 is the broader umbrella change, spanning cache/modeling/runtime support in addition to the heterogeneity scaffolding, so it is the better representative of the cluster.", "best_issue_reason": null, - "best_pr_reason": "PR 45333 is the cleaner representative of the pair because it establishes the foundational heterogeneous config API without the larger modeling surface area.", + "best_pr_reason": "PR 45332 is the strongest overall representative because it covers the larger end-to-end heterogeneous model support path, not just config plumbing.", "soft_edge_verdicts": [ { "left": "pull_request:45332", "right": "pull_request:45333", "accept": false, - "reason": "Related feature work, but not the same change: 45333 adds per-layer config support, while 45332 adds heterogeneous model/caching behavior and broader modeling integration. They could be separate PRs and are not duplicates." + "reason": "Related but distinct: 45333 adds per-layer configuration plumbing, while 45332 adds heterogeneous model/cache/runtime behavior. They share groundwork but not the same concrete change and would not be merged as one PR." } ] }, @@ -19758,12 +19803,12 @@ "soft_pairs": [ "pull_request:45415|pull_request:45425" ], - "prepared_review_unit_hash": "d3834e67f94a74a851d3f60ca5a51075186f84134bab55529de6962a22dd961c", + "prepared_review_unit_hash": "f40f6aafc379625350f465d38ba32d1581ad5377fce1c15d1057f27857b8dcab", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5514, + "serialized_chars": 5513, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -19771,7 +19816,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 5514, + "serialized_chars": 5513, "estimated_input_tokens": 1379, "estimated_eval_tokens": 3014 }, @@ -19780,18 +19825,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both pull requests are typing-related and overlap in a few files, but they are not the same change: #45415 is a broad type-checking sweep across many `src/transformers/*` modules plus `pyproject.toml`, while #45425 adds/adjusts typing helper protocols and narrower fixes around `_typing`, `modeling_utils`, and related utilities.", + "summary": "These are both typing-related PRs in transformers, but they address different levels of the problem: one is a broad type-checking sweep across many files, while the other adds specific typing protocols/aliases to support particular model utility code. They are not the same underlying change.", "confidence": 0.88, "canonical_issue_reason": null, - "canonical_pr_reason": "#45415 is the better canonical PR because it is the broader umbrella type-checking pass and touches the main set of files in the cluster; #45425 looks like a narrower follow-up focused on typing helper definitions.", + "canonical_pr_reason": "PR #45425 is the more focused, concrete change: it introduces shared typing aliases/protocols and touches a small, coherent set of type-related call sites.", "best_issue_reason": null, - "best_pr_reason": "#45415 best represents the cluster overall since it is the largest and most central typing cleanup, whereas #45425 is more specialized and dependency-like.", + "best_pr_reason": "PR #45425 is the better representative of a single underlying code change; PR #45415 is a broad housekeeping sweep that spans many unrelated typing cleanups.", "soft_edge_verdicts": [ { "left": "pull_request:45415", "right": "pull_request:45425", "accept": false, - "reason": "Shared typing theme and a few overlapping files are not enough; they do not appear to fix the same concrete code-path problem and would not plausibly be merged as one PR." + "reason": "Both are typing-related, but #45415 is a broad type-checking cleanup across many files while #45425 adds specific typing helpers/protocols for concrete code paths. They do not look mergeable as one PR or as the same fix." } ] }, @@ -19813,48 +19858,79 @@ "source": "llm", "cache_hit": false, "model": "gpt-5.4-mini?service_tier=flex", - "cluster_id": "cluster-45487-2", + "cluster_id": "cluster-45487-5", "nodes": [ "pull_request:45487", - "pull_request:46101" + "pull_request:46101", + "pull_request:46109", + "pull_request:46135", + "pull_request:46151" ], "soft_pairs": [ - "pull_request:45487|pull_request:46101" + "pull_request:45487|pull_request:46101", + "pull_request:45487|pull_request:46151", + "pull_request:46101|pull_request:46151", + "pull_request:46109|pull_request:46151", + "pull_request:46135|pull_request:46151" ], - "prepared_review_unit_hash": "273e1386555dd1fc8f90bff850866f4891fa7384f8eddb3f3c9b5a10993bad41", + "prepared_review_unit_hash": "0e268ef41b85b2782fe02c6be03fd2d57c002a5c4ebf1f1aea64ffbf3dfe6c90", "component_budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 5132, - "estimated_input_tokens": 1283, - "estimated_eval_tokens": 2822 + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 12529, + "estimated_input_tokens": 3133, + "estimated_eval_tokens": 6522 }, "budget": { - "node_count": 2, - "item_count": 2, - "soft_pair_count": 1, - "serialized_chars": 5132, - "estimated_input_tokens": 1283, - "estimated_eval_tokens": 2822 + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 12529, + "estimated_input_tokens": 3133, + "estimated_eval_tokens": 6522 }, "overflow_policy": "truncate_then_skip", "trimmed": false, "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "The two pull requests are both device/parallelism fixes, but they target different model families and different code paths, so they are not duplicates.", - "confidence": 0.97, + "summary": "This cluster is not a duplicate set: it mixes a couple of model-parallel device-placement fixes with unrelated test-only expectation updates. The closest items are thematically similar, but they target different models or test suites and should not be merged as one PR.", + "confidence": 0.91, "canonical_issue_reason": null, - "canonical_pr_reason": null, + "canonical_pr_reason": "PR 45487 is the strongest representative of the cluster because it fixes a concrete model-parallel device-placement bug in real model code and spans multiple affected model implementations plus tests.", "best_issue_reason": null, - "best_pr_reason": "PR 45487 is the more representative fix in this cluster: it has the broader, clearer bug description and addresses the AltCLIP/ChineseCLIP model-parallel path across multiple files.", + "best_pr_reason": "PR 45487 is the best standalone PR in the group: it addresses an actual runtime bug in model code, not just test expectations, and is broader/more substantive than the other candidates.", "soft_edge_verdicts": [ { "left": "pull_request:45487", "right": "pull_request:46101", "accept": false, - "reason": "Reject: these are unrelated fixes. 45487 patches token_type_ids device placement in AltCLIP/ChineseCLIP-related model-parallel code, while 46101 fixes audio feature masking device placement in Qwen2-Audio. Same general theme, different underlying bugs and code paths." + "reason": "Both are device-placement fixes for parallel execution, but they affect different model code paths (AltCLIP/ChineseCLIP vs Qwen2 Audio) and are not the same concrete bug." + }, + { + "left": "pull_request:45487", + "right": "pull_request:46151", + "accept": false, + "reason": "One is a model-parallel runtime fix; the other only adds XPU test expectations for token classification. Different artifacts and different underlying change." + }, + { + "left": "pull_request:46101", + "right": "pull_request:46151", + "accept": false, + "reason": "Qwen2 Audio parallel fix vs pipeline test expectation updates for token classification; same broad theme of device support, but unrelated bugs/changes." + }, + { + "left": "pull_request:46109", + "right": "pull_request:46151", + "accept": false, + "reason": "Distributed trainer test genericization is unrelated to token classification expectation updates; they do not fix the same issue or code path." + }, + { + "left": "pull_request:46135", + "right": "pull_request:46151", + "accept": false, + "reason": "Both touch XPU-related tests, but for different suites and behaviors (Kosmos2.5 integration vs token classification). Not the same underlying change." } ] }, @@ -19886,7 +19962,7 @@ "pull_request:45513|pull_request:46084", "pull_request:46084|pull_request:46098" ], - "prepared_review_unit_hash": "dbabc90cba684aea7cc989b8c89906e5b9bd9260f648d8d628b6b90bff0556d6", + "prepared_review_unit_hash": "5aafab669c6aa22fbddf6e5292c02045fc7f4e6a3540b3b9a8cd15b765411833", "component_budget": { "node_count": 3, "item_count": 3, @@ -19908,30 +19984,30 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "This cluster contains one unrelated PR (Qwen3.5/OLMo-Hybrid cache handling) and two closely related Mamba2 fixes for the same use_cache seq_len>1 bug. The broader Mamba2 PR 46098 is the best canonical representative.", - "confidence": 0.88, + "summary": "The two Mamba2 pull requests target the same cached-forward seq_len>1 bug, while the Qwen3.5/OLMo hybrid PR is a different model and code path.", + "confidence": 0.94, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 46098 is the most complete Mamba2 fix: it explicitly targets issue #46032, covers both CPU and GPU paths, and adds tests/docs, making it the cleanest canonical representative for the underlying bug.", + "canonical_pr_reason": "PR #46084 is the core implementation fix for Mamba2 use_cache with seq_len>1, with code and test changes directly addressing the incorrect cached forward path.", "best_issue_reason": null, - "best_pr_reason": "PR 46098 is the strongest overall choice because it fixes the same Mamba2 seq_len>1 cache bug as 46084 but does so more comprehensively and with clearer issue linkage.", + "best_pr_reason": "PR #46084 is the best representative of the underlying bug because it is the focused code-level fix, rather than a docs-heavy follow-up.", "soft_edge_verdicts": [ { "left": "pull_request:45513", "right": "pull_request:46084", "accept": false, - "reason": "Different models and bugs: 45513 is about Qwen3.5/OLMo-Hybrid cached forward behavior, while 46084 is a Mamba2 cache bug. The similarity is only at a generic pattern level." + "reason": "Different model family and bug: Qwen3.5/OLMO hybrid linear-attention caching versus Mamba2 cached forward behavior." }, { "left": "pull_request:46084", "right": "pull_request:46098", "accept": true, - "reason": "Both address the same Mamba2 use_cache seq_len>1 cached-forward bug in the same codebase and could plausibly be merged into one PR; 46098 is the broader superset fix." + "reason": "Same underlying Mamba2 use_cache seq_len>1 bug, same issue target, and overlapping modeling/test files; one could plausibly subsume the other in a single PR." } ] }, "evaluator_result": { "accept": true, - "feedback": "The cluster summary is grounded: 45513 is about Qwen/OLMo-Hybrid cached forward behavior, while 46084 and 46098 are both Mamba2 fixes tied to issue #46032 and appear mergeable as one bugfix PR. The soft-edge decisions are conservative, especially the rejection of 45513 vs 46084. The only slightly interpretive wording is the comparison between 46084 and 46098, but it is still supported by the packet." + "feedback": "The summary is grounded in the packet: PR #45513 is a different model/code path (Qwen3.5/OLMO hybrid linear-attention caching), while PRs #46084 and #46098 both address the same Mamba2 use_cache seq_len>1 bug and share the same issue target/files. The soft-edge verdicts are conservative and the mergeability judgment for the two Mamba2 PRs is plausible." }, "evaluator_used": true, "retried": false, @@ -19958,7 +20034,7 @@ "soft_pairs": [ "pull_request:45548|pull_request:45662" ], - "prepared_review_unit_hash": "cf201c1a9153f3024b111e79e00e6135674431573b29b85de4dde9b8c1a194fe", + "prepared_review_unit_hash": "24dd820d08e706b90c2bfffa45c564b6f2b1d0486588a40f4be3703bf223d7e8", "component_budget": { "node_count": 2, "item_count": 2, @@ -19980,18 +20056,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs are about expert parallelism, but they fix different concrete failure modes in different loading paths. They should not be merged as duplicates.", - "confidence": 0.95, + "summary": "These are related only at a high level (expert parallelism / model loading) but they fix different bugs in different code paths, so they should not be clustered as duplicates.", + "confidence": 0.93, "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45662 is the more concrete, code-path-specific EP fix: it addresses expert parameters being overwritten during loading/broadcast and patches the relevant loading/integration path directly.", + "canonical_pr_reason": "PR #45662 is the stronger canonical representative because it targets a concrete loading-time corruption bug (rank-0 broadcast overwriting experts) with a focused code-path fix across the relevant loading/integration files.", "best_issue_reason": null, - "best_pr_reason": "PR 45662 is the best representative of the cluster because it targets a specific bug with a clear mechanism and cohesive changes across the affected EP loading flow.", + "best_pr_reason": "PR #45662 is the better standalone representative of the cluster because it addresses a specific, reproducible EP/FSDP2 failure mode rather than a more general serialization/property cleanup.", "soft_edge_verdicts": [ { "left": "pull_request:45548", "right": "pull_request:45662", "accept": false, - "reason": "They share the broad EP theme, but the bugs differ: 45548 is about serializing distributed_config and adding has_ep for EP + ZeRO-3/accelerate launch loading, while 45662 fixes experts being overwritten by rank-0 broadcast in EP + FSDP2. Different root causes and code paths." + "reason": "Both mention EP and touch modeling_utils, but #45548 fixes config serialization and an EP capability flag for accelerate/ZeRO-3 loading, while #45662 fixes expert tensors being overwritten during FSDP2 loading. Different concrete bugs and merge paths." } ] }, @@ -20021,7 +20097,7 @@ "soft_pairs": [ "pull_request:45639|pull_request:45645" ], - "prepared_review_unit_hash": "b6760f9b4418c9a7e9a353ac0695788524aa82f2f4cb0ae28f867827fdb281f5", + "prepared_review_unit_hash": "eb845983aa4cc969553dc242db9412983f355a29a866bd19b33a9c299a278439", "component_budget": { "node_count": 2, "item_count": 2, @@ -20043,18 +20119,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Both PRs address the same xdist/captured_info logging collision in transformers testing utilities, with 45645 appearing to be the broader superset that also updates CI log collection. They are close enough to treat as duplicate candidates for the same underlying fix.", - "confidence": 0.84, - "canonical_issue_reason": null, - "canonical_pr_reason": "PR 45645 is the more complete implementation: it includes the xdist-safe testing_utils change plus the CI workflow update to collect multiple captured_info files, so it better represents the full fix path.", + "summary": "Both PRs address the same xdist-related captured_info logging problem. PR 45645 is the more complete change because it includes the testing_utils fix plus the CI log aggregation updates, while 45639 is just the core code-side piece.", + "confidence": 0.96, + "canonical_issue_reason": "Both PRs explicitly target issue #45561, which appears to be the shared tracking issue for the xdist-safe captured_info fix.", + "canonical_pr_reason": "PR 45645 is the best canonical PR because it subsumes the core testing_utils change from 45639 and also updates the workflow to read multiple captured_info files safely in CI.", "best_issue_reason": null, - "best_pr_reason": "PR 45645 is the best representative because it covers the core bug fix and the downstream CI/debug-log handling needed to preserve the new per-worker artifacts.", + "best_pr_reason": "PR 45645 is the strongest representative of the change set: it fixes the xdist collision in code and preserves debug-log collection in the workflow, whereas 45639 only covers the code-path portion.", "soft_edge_verdicts": [ { "left": "pull_request:45639", "right": "pull_request:45645", "accept": true, - "reason": "Same underlying xdist-safe captured_info logging change in the same code path; 45645 looks like a superset of 45639 with added CI handling, so they can plausibly be merged into one PR." + "reason": "Same underlying xdist-safe captured_info bug; 45645 extends the same fix rather than addressing a different change." } ] }, @@ -20150,12 +20226,12 @@ "soft_pairs": [ "pull_request:46069|pull_request:46073" ], - "prepared_review_unit_hash": "a1f231e419f7cfd6aa85433cd57eafeb203a51f245805c7101b76f3ddd813f1e", + "prepared_review_unit_hash": "71120b3c8458907102557418114771428787fce6f13761d0077f90f720cc52e5", "component_budget": { "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4394, + "serialized_chars": 4393, "estimated_input_tokens": 1099, "estimated_eval_tokens": 2454 }, @@ -20163,7 +20239,7 @@ "node_count": 2, "item_count": 2, "soft_pair_count": 1, - "serialized_chars": 4394, + "serialized_chars": 4393, "estimated_input_tokens": 1099, "estimated_eval_tokens": 2454 }, @@ -20172,18 +20248,18 @@ "aggressively_trimmed": false, "split": false, "analyst_result": { - "summary": "Two separate model-specific PRs adding SDPA/attn_implementation dispatch support. They share a theme but not the same underlying code path or mergeable change, so they should not be clustered as duplicates.", - "confidence": 0.96, + "summary": "These are two unrelated model-specific PRs adding attention-dispatch support to different architectures (ShieldGemma2 vs CTRL), so they should not be clustered as duplicates.", + "confidence": 0.98, "canonical_issue_reason": null, "canonical_pr_reason": null, "best_issue_reason": null, - "best_pr_reason": "No single PR is a clear canonical representative: one targets ShieldGemma2 and the other CTRL, with different implementations and files touched.", + "best_pr_reason": "If a representative PR is needed, #46069 is the more substantial one (code, tests, and docs), but it is not the same change as #46073.", "soft_edge_verdicts": [ { "left": "pull_request:46069", "right": "pull_request:46073", "accept": false, - "reason": "Both are about attention dispatch support, but they apply to different model architectures and separate modeling files; they are not the same concrete bug/change and would not be merged into one PR." + "reason": "Different models and different concrete implementations: ShieldGemma2 support vs CTRL sdpa dispatch. They do not fix the same code-path problem and would not plausibly merge into one PR." } ] }, diff --git a/analysis/current/manifest.json b/analysis/current/manifest.json index 81647ce98b77e28cd3882fefde7266ae7b3aedc0..db08fadd17c35ad557249e85574a57612d433818 100644 --- a/analysis/current/manifest.json +++ b/analysis/current/manifest.json @@ -1,8 +1,8 @@ { - "analysis_id": "hybrid-model-20260522t000112z", + "analysis_id": "hybrid-model-20260522t060021z", "archived_artifacts": { - "hybrid": "snapshots/20260522T000112Z/analysis-runs/hybrid-model-20260522t000112z/analysis-report-hybrid.json", - "hybrid_reviews": "snapshots/20260522T000112Z/analysis-runs/hybrid-model-20260522t000112z/analysis-report-hybrid.llm-reviews.json" + "hybrid": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.llm-reviews.json" }, "artifacts": { "hybrid": "analysis/current/analysis-report-hybrid.json", @@ -10,9 +10,9 @@ }, "channel": "canonical", "model": null, - "published_at": "2026-05-22T00:05:47Z", + "published_at": "2026-05-22T06:10:05Z", "repo": "huggingface/transformers", "schema_version": 1, - "snapshot_id": "20260522T000112Z", + "snapshot_id": "20260522T060021Z", "variant": "hybrid" } diff --git a/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.json b/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.json new file mode 100644 index 0000000000000000000000000000000000000000..f5d79c384b7a6be49454400baf53c28196480091 --- /dev/null +++ b/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.json @@ -0,0 +1,2156 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260522T060021Z", + "generated_at": "2026-05-22T06:10:04Z", + "evidence_quality": "full", + "llm_enrichment": true, + "meta_bugs": [ + { + "cluster_id": "cluster-43979-11", + "summary": "Cluster of 1 issues and 10 PRs centered on issue #43979.", + "status": "open", + "confidence": 0.8, + "canonical_issue_number": 43979, + "canonical_pr_number": 44007, + "issue_numbers": [ + 43979 + ], + "pr_numbers": [ + 43996, + 44007, + 44013, + 44044, + 44066, + 44072, + 44085, + 44129, + 44154, + 44722 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target", + "soft_similarity" + ], + "pr_comparisons": [ + { + "left_pr_number": 43996, + "right_pr_number": 44007, + "code_similarity": 0.179, + "size_similarity": 0.576, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.429, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44013, + "code_similarity": 0.122, + "size_similarity": 0.318, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.392, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44044, + "code_similarity": 0.245, + "size_similarity": 0.864, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.479, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44066, + "code_similarity": 0.225, + "size_similarity": 0.818, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.408, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44072, + "code_similarity": 0.14, + "size_similarity": 0.303, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.528, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44085, + "code_similarity": 0.216, + "size_similarity": 0.783, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44129, + "code_similarity": 0.163, + "size_similarity": 0.643, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.229, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44154, + "code_similarity": 0.153, + "size_similarity": 0.535, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.31, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 43996, + "right_pr_number": 44722, + "code_similarity": 0.225, + "size_similarity": 0.848, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.368, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44013, + "code_similarity": 0.19, + "size_similarity": 0.553, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.531, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44044, + "code_similarity": 0.186, + "size_similarity": 0.667, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.354, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44066, + "code_similarity": 0.188, + "size_similarity": 0.704, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.315, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44072, + "code_similarity": 0.212, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.708, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44085, + "code_similarity": 0.195, + "size_similarity": 0.735, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.318, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44129, + "code_similarity": 0.103, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.191, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44154, + "code_similarity": 0.238, + "size_similarity": 0.93, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.344, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44007, + "right_pr_number": 44722, + "code_similarity": 0.178, + "size_similarity": 0.679, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.28, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44044, + "code_similarity": 0.126, + "size_similarity": 0.368, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.351, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44066, + "code_similarity": 0.127, + "size_similarity": 0.389, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.325, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44072, + "code_similarity": 0.29, + "size_similarity": 0.952, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.667, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44085, + "code_similarity": 0.131, + "size_similarity": 0.406, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.329, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44129, + "code_similarity": 0.07, + "size_similarity": 0.205, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.192, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44154, + "code_similarity": 0.177, + "size_similarity": 0.594, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.389, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44013, + "right_pr_number": 44722, + "code_similarity": 0.118, + "size_similarity": 0.375, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.287, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44066, + "code_similarity": 0.25, + "size_similarity": 0.947, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.404, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44072, + "code_similarity": 0.136, + "size_similarity": 0.351, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.442, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44085, + "code_similarity": 0.24, + "size_similarity": 0.906, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.394, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44129, + "code_similarity": 0.147, + "size_similarity": 0.555, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.243, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44154, + "code_similarity": 0.17, + "size_similarity": 0.62, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.306, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44044, + "right_pr_number": 44722, + "code_similarity": 0.257, + "size_similarity": 0.982, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.402, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44072, + "code_similarity": 0.133, + "size_similarity": 0.37, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.393, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44085, + "code_similarity": 0.763, + "size_similarity": 0.957, + "file_overlap": 0.5, + "area_overlap": 0.825, + "patch_similarity": 0.887, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ] + } + ] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44129, + "code_similarity": 0.145, + "size_similarity": 0.526, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.263, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44154, + "code_similarity": 0.174, + "size_similarity": 0.654, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.286, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44066, + "right_pr_number": 44722, + "code_similarity": 0.907, + "size_similarity": 0.964, + "file_overlap": 1.0, + "area_overlap": 0.808, + "patch_similarity": 0.874, + "shared_filenames": [ + "src/transformers/models/codegen/modeling_codegen.py", + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/codegen/modeling_codegen.py", + "left_ranges": [ + [ + 245, + 266 + ] + ], + "right_ranges": [ + [ + 228, + 234 + ] + ] + }, + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 41 + ], + [ + 174, + 181 + ], + [ + 250, + 257 + ], + [ + 398, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 466 + ], + [ + 471, + 488 + ], + [ + 518, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 606 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44085, + "code_similarity": 0.137, + "size_similarity": 0.387, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.398, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44129, + "code_similarity": 0.074, + "size_similarity": 0.195, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.231, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44154, + "code_similarity": 0.175, + "size_similarity": 0.566, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.414, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44072, + "right_pr_number": 44722, + "code_similarity": 0.124, + "size_similarity": 0.357, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.347, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44129, + "code_similarity": 0.141, + "size_similarity": 0.503, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.272, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44154, + "code_similarity": 0.18, + "size_similarity": 0.684, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.289, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44085, + "right_pr_number": 44722, + "code_similarity": 0.728, + "size_similarity": 0.923, + "file_overlap": 0.5, + "area_overlap": 0.791, + "patch_similarity": 0.78, + "shared_filenames": [ + "src/transformers/models/gptj/modeling_gptj.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/models/gptj/modeling_gptj.py", + "left_ranges": [ + [ + 33, + 40 + ], + [ + 173, + 180 + ], + [ + 249, + 256 + ], + [ + 397, + 405 + ], + [ + 408, + 419 + ], + [ + 425, + 434 + ], + [ + 459, + 465 + ], + [ + 470, + 489 + ], + [ + 517, + 539 + ], + [ + 553, + 559 + ], + [ + 565, + 574 + ], + [ + 579, + 597 + ], + [ + 600, + 611 + ], + [ + 633, + 639 + ], + [ + 645, + 652 + ], + [ + 657, + 673 + ], + [ + 716, + 728 + ], + [ + 737, + 743 + ], + [ + 748, + 771 + ], + [ + 789, + 794 + ] + ], + "right_ranges": [ + [ + 33, + 42 + ], + [ + 175, + 182 + ], + [ + 251, + 258 + ], + [ + 399, + 420 + ], + [ + 426, + 435 + ], + [ + 460, + 467 + ], + [ + 472, + 489 + ], + [ + 519, + 540 + ], + [ + 554, + 560 + ], + [ + 566, + 575 + ], + [ + 580, + 598 + ], + [ + 601, + 612 + ], + [ + 634, + 640 + ], + [ + 646, + 653 + ], + [ + 658, + 663 + ], + [ + 666, + 674 + ], + [ + 717, + 722 + ], + [ + 738, + 744 + ], + [ + 749, + 772 + ], + [ + 790, + 795 + ] + ] + } + ] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44154, + "code_similarity": 0.099, + "size_similarity": 0.344, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.199, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44129, + "right_pr_number": 44722, + "code_similarity": 0.146, + "size_similarity": 0.545, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.247, + "shared_filenames": [], + "shared_file_areas": [] + }, + { + "left_pr_number": 44154, + "right_pr_number": 44722, + "code_similarity": 0.164, + "size_similarity": 0.631, + "file_overlap": 0.0, + "area_overlap": 0.0, + "patch_similarity": 0.255, + "shared_filenames": [], + "shared_file_areas": [] + } + ] + }, + { + "cluster_id": "cluster-41211-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41211.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41211, + "canonical_pr_number": 44339, + "issue_numbers": [ + 41211 + ], + "pr_numbers": [ + 41356, + 44339 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41356, + "right_pr_number": 44339, + "code_similarity": 0.155, + "size_similarity": 0.094, + "file_overlap": 0.273, + "area_overlap": 0.078, + "patch_similarity": 0.183, + "shared_filenames": [ + "docs/source/en/model_doc/deimv2.md", + "src/transformers/models/deimv2/__init__.py", + "src/transformers/models/deimv2/configuration_deimv2.py", + "src/transformers/models/deimv2/modeling_deimv2.py", + "tests/models/deimv2/__init__.py", + "tests/models/deimv2/test_modeling_deimv2.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/model_doc/deimv2.md", + "left_ranges": [ + [ + 1, + 132 + ] + ], + "right_ranges": [ + [ + 1, + 65 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/__init__.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/deimv2/configuration_deimv2.py", + "left_ranges": [ + [ + 1, + 74 + ] + ], + "right_ranges": [ + [ + 1, + 266 + ] + ] + }, + { + "filename": "tests/models/deimv2/test_modeling_deimv2.py", + "left_ranges": [ + [ + 1, + 15 + ] + ], + "right_ranges": [ + [ + 1, + 1734 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-43240-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #43240.", + "status": "open", + "confidence": 0.75, + "canonical_issue_number": 43240, + "canonical_pr_number": 43251, + "issue_numbers": [ + 43240 + ], + "pr_numbers": [ + 43251, + 43254 + ], + "evidence_types": [ + "closing_reference", + "shared_issue_target" + ], + "pr_comparisons": [ + { + "left_pr_number": 43251, + "right_pr_number": 43254, + "code_similarity": 0.794, + "size_similarity": 0.64, + "file_overlap": 1.0, + "area_overlap": 0.667, + "patch_similarity": 0.882, + "shared_filenames": [ + "src/transformers/loss/loss_utils.py" + ], + "shared_file_areas": [ + { + "filename": "src/transformers/loss/loss_utils.py", + "left_ranges": [ + [ + 30, + 50 + ] + ], + "right_ranges": [ + [ + 30, + 43 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-41115-3", + "summary": "Cluster of 1 issues and 2 PRs centered on issue #41115.", + "status": "open", + "confidence": 0.55, + "canonical_issue_number": 41115, + "canonical_pr_number": 45613, + "issue_numbers": [ + 41115 + ], + "pr_numbers": [ + 41116, + 45613 + ], + "evidence_types": [ + "closing_reference" + ], + "pr_comparisons": [ + { + "left_pr_number": 41116, + "right_pr_number": 45613, + "code_similarity": 0.805, + "size_similarity": 0.885, + "file_overlap": 0.75, + "area_overlap": 0.854, + "patch_similarity": 0.693, + "shared_filenames": [ + "docs/source/en/_toctree.yml", + "docs/source/en/model_doc/minicpm3.md", + "src/transformers/models/auto/modeling_auto.py", + "src/transformers/models/minicpm3/__init__.py", + "src/transformers/models/minicpm3/configuration_minicpm3.py", + "src/transformers/models/minicpm3/modeling_minicpm3.py", + "src/transformers/models/minicpm3/modular_minicpm3.py", + "tests/models/minicpm3/__init__.py", + "tests/models/minicpm3/test_modeling_minicpm3.py" + ], + "shared_file_areas": [ + { + "filename": "docs/source/en/_toctree.yml", + "left_ranges": [ + [ + 713, + 720 + ] + ], + "right_ranges": [ + [ + 713, + 720 + ] + ] + }, + { + "filename": "docs/source/en/model_doc/minicpm3.md", + "left_ranges": [ + [ + 1, + 67 + ] + ], + "right_ranges": [ + [ + 1, + 45 + ] + ] + }, + { + "filename": "src/transformers/models/auto/modeling_auto.py", + "left_ranges": [ + [ + 285, + 291 + ], + [ + 700, + 706 + ], + [ + 1298, + 1304 + ] + ], + "right_ranges": [ + [ + 699, + 705 + ], + [ + 1301, + 1307 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/__init__.py", + "left_ranges": [ + [ + 1, + 29 + ] + ], + "right_ranges": [ + [ + 1, + 29 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/configuration_minicpm3.py", + "left_ranges": [ + [ + 1, + 141 + ] + ], + "right_ranges": [ + [ + 1, + 126 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 544 + ] + ], + "right_ranges": [ + [ + 1, + 522 + ] + ] + }, + { + "filename": "src/transformers/models/minicpm3/modular_minicpm3.py", + "left_ranges": [ + [ + 1, + 444 + ] + ], + "right_ranges": [ + [ + 1, + 342 + ] + ] + }, + { + "filename": "tests/models/minicpm3/test_modeling_minicpm3.py", + "left_ranges": [ + [ + 1, + 109 + ] + ], + "right_ranges": [ + [ + 1, + 136 + ] + ] + } + ] + } + ] + }, + { + "cluster_id": "cluster-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-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 12 discussion comments plus 8 inbound references.", + "score": 162.0 + }, + "best_pr": { + "cluster_id": "cluster-43979-11", + "pr_number": 44007, + "reason": "PR #44007 is the strongest global PR candidate because it is open, belongs to a cluster with 11 artifacts, links to 1 issues, and carries 2 review/discussion events.", + "score": 238.886 + } +} diff --git a/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.llm-reviews.json b/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.llm-reviews.json new file mode 100644 index 0000000000000000000000000000000000000000..069af027274340ffbb3ab609b4006a8e68da95b8 --- /dev/null +++ b/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.llm-reviews.json @@ -0,0 +1,20274 @@ +{ + "schema_version": "1.0", + "repo": "huggingface/transformers", + "snapshot_id": "20260522T060021Z", + "generated_at": "2026-05-22T06:10:04Z", + "model": "gpt-5.4-mini?service_tier=flex", + "reviews": [ + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:24643", + "issue:30277", + "issue:34634", + "issue:34928", + "issue:35545", + "issue:36010", + "issue:36331", + "issue:39290", + "issue:41093", + "issue:41720", + "issue:41762", + "issue:42915", + "issue:43039", + "issue:43425", + "issue:43531", + "issue:43541", + "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" + ], + "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:43638|issue:44805", + "issue:43596|issue:44805", + "issue:44805|issue:45161", + "issue:44805|issue:44918", + "issue:34634|issue:35545", + "issue:30277|issue:43638", + "issue:43596|issue:43638", + "issue:43039|issue:45161", + "issue:43975|issue:44457", + "issue:44292|issue:44457", + "issue:34928|issue:44805", + "issue:24643|issue:30277", + "issue:45859|issue:45902", + "issue:43638|issue:45161", + "issue:34928|issue:36331", + "issue:43596|issue:45161", + "issue:43716|issue:45507", + "issue:39290|issue:45081", + "issue:45823|issue:45874", + "issue:43425|issue:44292", + "issue:34928|issue:41720", + "issue:43828|issue:45161", + "issue:44918|issue:45161", + "issue:43716|issue:45685", + "issue:43541|issue:43716", + "issue:43828|issue:44292", + "issue:45507|issue:45685", + "issue:43716|issue:44292", + "issue:44292|issue:44918", + "issue:43039|issue:44918", + "issue:41762|issue:43638", + "issue:43827|issue:44512", + "issue:43716|issue:44560", + "issue:43828|issue:45685", + "issue:44918|issue:45685", + "issue:43039|issue:43828", + "issue:30277|issue:36331", + "issue:45237|issue:46015", + "issue:34928|issue:36010", + "issue:43039|issue:44805", + "issue:41762|issue:43596", + "issue:43901|issue:44512", + "issue:39290|issue:41720", + "issue:45237|issue:45901", + "issue:44841|issue:45823", + "issue:43039|issue:44560", + "issue:43975|issue:46015", + "issue:45812|issue:45920", + "issue:44863|issue:46015", + "issue:43866|issue:44863", + "issue:44292|issue:46015", + "issue:39290|issue:43531", + "issue:41093|issue:43638" + ], + "prepared_review_unit_hash": "292d8f46e520f108716c719dfd9801e6aee773095874bda6d2225fe08bae8274", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 66, + "serialized_chars": 47536, + "estimated_input_tokens": 11884, + "estimated_eval_tokens": 24024 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous and the soft-similarity links are largely false positives. The only clear duplicate pockets are the DeepSpeed ZeRO-3 Bert index error pair and the AutoTokenizer v5 token-ID regression pair.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 45920 is the broadest, most representative report of the shared AutoTokenizer v5 token-ID regression, and it subsumes the narrower Granite-only report in 45812.", + "canonical_pr_reason": null, + "best_issue_reason": "45920 best captures the underlying bug because it describes the regression across multiple models, making it the most useful umbrella issue.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44322", + "right": "issue:45464", + "accept": false, + "reason": "Different failures: Qwen3.5 config attribute error vs streaming chat/completions failure." + }, + { + "left": "issue:45237", + "right": "issue:45685", + "accept": false, + "reason": "Unrelated model/runtime issues; AMD GPU support is not the same bug as the MPS histogram error." + }, + { + "left": "issue:43716", + "right": "issue:45237", + "accept": false, + "reason": "Both are device/dtype-related, but they affect different models and code paths." + }, + { + "left": "issue:43828", + "right": "issue:45237", + "accept": false, + "reason": "Autocast dtype mismatch and AMD GPU incompatibility are not the same underlying problem." + }, + { + "left": "issue:45237", + "right": "issue:45507", + "accept": false, + "reason": "Different model families and different failure modes." + }, + { + "left": "issue:44292", + "right": "issue:45237", + "accept": false, + "reason": "Qwen-3 NVFP4 failure is unrelated to the AMD GPU issue." + }, + { + "left": "issue:30277", + "right": "issue:35545", + "accept": false, + "reason": "DeepSpeed collective mismatch and ONNX export failure are separate bugs." + }, + { + "left": "issue:35545", + "right": "issue:42915", + "accept": false, + "reason": "Both are model-specific export/load failures, but not the same concrete code path." + }, + { + "left": "issue:35545", + "right": "issue:41720", + "accept": false, + "reason": "ONNX export and device-map cuda assert are unrelated." + }, + { + "left": "issue:45901", + "right": "issue:46015", + "accept": false, + "reason": "Table QA crash and outdated RAG docs are different issues." + }, + { + "left": "issue:44509", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs regressions in v5, but they cover different removed features/pages and are not one bug." + }, + { + "left": "issue:35545", + "right": "issue:36010", + "accept": false, + "reason": "ONNX export failure and GenerationMixin import failure are different breakages." + }, + { + "left": "issue:44805", + "right": "issue:45823", + "accept": false, + "reason": "Mask shape mismatch and Gemma4 PLE device mismatch are distinct runtime errors." + }, + { + "left": "issue:43638", + "right": "issue:44805", + "accept": false, + "reason": "Different error signatures and different model/setup contexts." + }, + { + "left": "issue:43596", + "right": "issue:44805", + "accept": false, + "reason": "DeepSpeed ZeRO-3 Bert init error is not the same as the mask-shape mismatch." + }, + { + "left": "issue:44805", + "right": "issue:45161", + "accept": false, + "reason": "Mask indexing failure and GPT-OSS TP failure are unrelated." + }, + { + "left": "issue:44805", + "right": "issue:44918", + "accept": false, + "reason": "Different models and different failures." + }, + { + "left": "issue:34634", + "right": "issue:35545", + "accept": false, + "reason": "BarkProcessor voice_preset and ModernBERT ONNX export are unrelated." + }, + { + "left": "issue:30277", + "right": "issue:43638", + "accept": false, + "reason": "Both involve DeepSpeed, but one is a collective mismatch and the other is a ZeRO-3 index error." + }, + { + "left": "issue:43596", + "right": "issue:43638", + "accept": true, + "reason": "Same DeepSpeed ZeRO-3 index-0-out-of-bounds failure on Bert models; these look like the same underlying bug." + }, + { + "left": "issue:43039", + "right": "issue:45161", + "accept": false, + "reason": "Liger Kernel cross-entropy being called and GPT-OSS TP not working are different issues." + }, + { + "left": "issue:43975", + "right": "issue:44457", + "accept": false, + "reason": "Detokenization regression and LoRA save/load mismatch are not the same bug." + }, + { + "left": "issue:44292", + "right": "issue:44457", + "accept": false, + "reason": "Qwen NVFP4 runtime failure and LoRA merge/save mismatch are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:44805", + "accept": false, + "reason": "Activation checkpointing/FSDP tensor recomputation mismatch is not the same as the mask-shape index error." + }, + { + "left": "issue:24643", + "right": "issue:30277", + "accept": false, + "reason": "Different DeepSpeed errors with different symptoms and code paths." + }, + { + "left": "issue:45859", + "right": "issue:45902", + "accept": false, + "reason": "Both are Qwen3.5 MoE rotary issues, but one is CPU offload compatibility and the other is uninitialized memory after to_empty materialization." + }, + { + "left": "issue:43638", + "right": "issue:45161", + "accept": false, + "reason": "DeepSpeed Bert index error and GPT-OSS TP failure are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36331", + "accept": false, + "reason": "Checkpointing/FSDP recomputation mismatch and custom trainer signature change are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:45161", + "accept": false, + "reason": "Bert ZeRO-3 index error is unrelated to GPT-OSS tensor-parallel issues." + }, + { + "left": "issue:43716", + "right": "issue:45507", + "accept": false, + "reason": "Different model families and failure modes." + }, + { + "left": "issue:39290", + "right": "issue:45081", + "accept": false, + "reason": "Gemma3/vLLM sliding-window attribute error and Mistral tokenizer regex crash are unrelated." + }, + { + "left": "issue:45823", + "right": "issue:45874", + "accept": false, + "reason": "Device-map forward mismatch and Gemma4 reverse embedding lookup are different problems." + }, + { + "left": "issue:43425", + "right": "issue:44292", + "accept": false, + "reason": "Torch version incompatibility and Qwen NVFP4 runtime failure are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:41720", + "accept": false, + "reason": "FSDP checkpointing mismatch and cuda assert under auto device mapping are different bugs." + }, + { + "left": "issue:43828", + "right": "issue:45161", + "accept": false, + "reason": "Autocast dtype mismatch and GPT-OSS TP failure are not the same code path." + }, + { + "left": "issue:44918", + "right": "issue:45161", + "accept": false, + "reason": "Qwen3.5 embedding unpacking with TRL and GPT-OSS TP failure are unrelated." + }, + { + "left": "issue:43716", + "right": "issue:45685", + "accept": false, + "reason": "Different model/runtime bugs." + }, + { + "left": "issue:43828", + "right": "issue:44292", + "accept": false, + "reason": "Autocast dtype mismatch and Qwen NVFP4 failure are different issues." + }, + { + "left": "issue:43039", + "right": "issue:44918", + "accept": false, + "reason": "Liger Kernel cross-entropy call and TRL embedding unpacking are different problems." + }, + { + "left": "issue:41762", + "right": "issue:43638", + "accept": false, + "reason": "Both are ZeRO-3 index errors, but they involve different models and likely different initialization paths." + }, + { + "left": "issue:43827", + "right": "issue:44512", + "accept": false, + "reason": "Both are docs regressions, but they reference different removed features and should not be merged as one duplicate." + }, + { + "left": "issue:43716", + "right": "issue:44560", + "accept": false, + "reason": "Different multimodal model failures and different symptoms." + }, + { + "left": "issue:43828", + "right": "issue:45685", + "accept": false, + "reason": "Autocast dtype mismatch and MPS Int histogram error are unrelated." + }, + { + "left": "issue:44918", + "right": "issue:45685", + "accept": false, + "reason": "TRL embedding unpacking and MPS histogram error are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:43828", + "accept": false, + "reason": "Liger Kernel cross-entropy path and autocast dtype mismatch are different bugs." + }, + { + "left": "issue:30277", + "right": "issue:36331", + "accept": false, + "reason": "DeepSpeed collective mismatch and custom trainer signature mismatch are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "AMD GPU support issue and table QA crash are unrelated." + }, + { + "left": "issue:34928", + "right": "issue:36010", + "accept": false, + "reason": "Activation checkpointing/FSDP mismatch and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:43039", + "right": "issue:44805", + "accept": false, + "reason": "Liger Kernel cross-entropy call and mask-shape mismatch are different bugs." + }, + { + "left": "issue:41762", + "right": "issue:43596", + "accept": false, + "reason": "Same broad DeepSpeed/ZeRO-3 area, but not enough evidence they are the same underlying defect." + }, + { + "left": "issue:43901", + "right": "issue:44512", + "accept": false, + "reason": "Docs mention different deprecated behaviors; not the same issue." + }, + { + "left": "issue:39290", + "right": "issue:41720", + "accept": false, + "reason": "vLLM sliding-window attribute error and cuda assert under auto device mapping are unrelated." + }, + { + "left": "issue:45237", + "right": "issue:45901", + "accept": false, + "reason": "Different subsystems and different failures." + }, + { + "left": "issue:44841", + "right": "issue:45823", + "accept": false, + "reason": "Processor failure for Voxtral and Gemma4 device mismatch are not the same bug." + }, + { + "left": "issue:43039", + "right": "issue:44560", + "accept": false, + "reason": "Liger Kernel cross-entropy issue and Qwen3-vl StopIteration are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:46015", + "accept": false, + "reason": "Detokenization bug and RAG docs update are unrelated." + }, + { + "left": "issue:45812", + "right": "issue:45920", + "accept": true, + "reason": "Same AutoTokenizer wrong-token-ID regression; 45920 is the broader multi-model report of the same bug." + }, + { + "left": "issue:44863", + "right": "issue:46015", + "accept": false, + "reason": "NemotronH checkpoint loading and outdated RAG examples are unrelated." + }, + { + "left": "issue:43866", + "right": "issue:44863", + "accept": false, + "reason": "Checkpoint corruption and NemotronH load failure are different issues." + }, + { + "left": "issue:44292", + "right": "issue:46015", + "accept": false, + "reason": "Qwen NVFP4 runtime failure and RAG docs are unrelated." + }, + { + "left": "issue:39290", + "right": "issue:43531", + "accept": false, + "reason": "Gemma3/vLLM sliding-window attribute error and Qwen3-MoE sliding_window issue are different model/config bugs." + }, + { + "left": "issue:41093", + "right": "issue:43638", + "accept": false, + "reason": "Similar IndexError wording, but different shapes, models, and contexts; not clearly the same bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet, and the two accepted duplicate pockets are well supported by the titles: DeepSpeed ZeRO-3 Bert index errors (43596/43638) and the AutoTokenizer v5 wrong-token-ID regression (45812/45920). The rejected links are generally conservative and do not overstate evidence, and the canonical issue choice of 45920 is justified as the broader umbrella report." + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:36010", + "issue:39186", + "issue:39290", + "issue:40990", + "issue:41093", + "issue:41762", + "issue:43054", + "issue:43296", + "issue:43366", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43541", + "issue:43572", + "issue:43596", + "issue:43645", + "issue:43716", + "issue:43792", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43975", + "issue:43994", + "issue:44291", + "issue:44292", + "issue:44387", + "issue:44530", + "issue:44589", + "issue:44841", + "issue:44918", + "issue:44960", + "issue:45005", + "issue:45020", + "issue:45070", + "issue:45084", + "issue:45161", + "issue:45362", + "issue:45399", + "issue:45464", + "issue:45542", + "issue:45632", + "issue:45685", + "issue:45698", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45901", + "issue:45902" + ], + "soft_pairs": [ + "issue:43854|issue:43866", + "issue:43792|issue:45901", + "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:43572|issue:45070", + "issue:43296|issue:43366", + "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:45542|issue:45800", + "issue:45759|issue:45902", + "issue:39186|issue:44918", + "issue:43716|issue:44918", + "issue:40990|issue:44960", + "issue:45632|issue:45698", + "issue:43645|issue:45399", + "issue:43404|issue:45005", + "issue:44530|issue:45005" + ], + "prepared_review_unit_hash": "81f617f1c961dc0df36ede00d0176fd04ff5c68eb3cfbc116646d436aeff94a4", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 38, + "serialized_chars": 38793, + "estimated_input_tokens": 9699, + "estimated_eval_tokens": 19654 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a collection of distinct model/runtime regression reports rather than one duplicate cluster. I only found one plausible duplicate pair around Siglip2 output quality; the rest are too model- or path-specific to merge confidently.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45020 is the broadest and most central report in the set, capturing the recurring 'recent transformers versions break model loading/behavior' theme better than the narrower model-specific bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45020 is the best cluster representative because it is the most general compatibility/regression issue and most likely to subsume related follow-up reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43854", + "right": "issue:43866", + "accept": false, + "reason": "Different models and failures: GLM-4.7-Flash loading in tests vs Ovis2 checkpoint corruption." + }, + { + "left": "issue:43792", + "right": "issue:45901", + "accept": false, + "reason": "Whisper inference failure and table-question-answering crash are unrelated code paths." + }, + { + "left": "issue:43425", + "right": "issue:45070", + "accept": false, + "reason": "Torch 2.10 compatibility and a PretrainedConfig pydantic field regression are different bugs." + }, + { + "left": "issue:43596", + "right": "issue:45800", + "accept": false, + "reason": "DeepSpeed ZeRO-3/BertModel index error is not the same as a torch 2.4.1 vs transformers 5.8 incompatibility." + }, + { + "left": "issue:43975", + "right": "issue:45901", + "accept": false, + "reason": "Incorrect detokenization in v5 and table QA crashes are unrelated." + }, + { + "left": "issue:43792", + "right": "issue:45070", + "accept": false, + "reason": "Different failures with no shared concrete code path." + }, + { + "left": "issue:45362", + "right": "issue:45464", + "accept": false, + "reason": "Both mention Qwen3.5, but one is a chat crash and the other is a streaming inference API failure on a different model size/path." + }, + { + "left": "issue:44292", + "right": "issue:45901", + "accept": false, + "reason": "Unrelated model/runtime error and table QA crash." + }, + { + "left": "issue:39290", + "right": "issue:45823", + "accept": false, + "reason": "Gemma3 sliding-window attribute error and Gemma4 device mismatch are different regressions." + }, + { + "left": "issue:45464", + "right": "issue:45685", + "accept": false, + "reason": "Qwen3.5 streaming failure and an MPS histogram kernel limitation are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:43866", + "accept": false, + "reason": "Weight-tying bug 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 not the same as the MPS Int histogram error." + }, + { + "left": "issue:43541", + "right": "issue:43828", + "accept": false, + "reason": "Both involve runtime numeric issues, but they are different models and different failure mechanisms." + }, + { + "left": "issue:44841", + "right": "issue:45084", + "accept": false, + "reason": "Voxtral processor failure and a template-node compile TypeError are unrelated." + }, + { + "left": "issue:36010", + "right": "issue:39290", + "accept": false, + "reason": "ImportError for GenerationMixin and a Gemma3 config attribute error are different bugs." + }, + { + "left": "issue:45800", + "right": "issue:45823", + "accept": false, + "reason": "Generic torch/transformers incompatibility and Gemma4 device placement mismatch are not the same issue." + }, + { + "left": "issue:41093", + "right": "issue:41762", + "accept": false, + "reason": "Both are index-related, but one is a mask shape mismatch and the other is a ZeRO-3 load-time out-of-bounds error." + }, + { + "left": "issue:43572", + "right": "issue:45070", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig and a pydantic field regression are different breakages." + }, + { + "left": "issue:43296", + "right": "issue:43366", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM and GGUF gpt-oss support are different feature/issues." + }, + { + "left": "issue:43541", + "right": "issue:45161", + "accept": false, + "reason": "Mixtral torch-dynamo tracing error and GPT-OSS tensor parallelism failure are different code paths." + }, + { + "left": "issue:44387", + "right": "issue:45005", + "accept": false, + "reason": "Int4 OOM regression and tied-weights translation-model issue are unrelated." + }, + { + "left": "issue:43828", + "right": "issue:45464", + "accept": false, + "reason": "Phi-tiny-MoE autocast dtype mismatch and Qwen3.5 streaming API failure are different bugs." + }, + { + "left": "issue:44291", + "right": "issue:45070", + "accept": false, + "reason": "init_empty_weights/_is_hf_initialized argument handling is unrelated to the pydantic model breakage." + }, + { + "left": "issue:43366", + "right": "issue:45464", + "accept": false, + "reason": "GGUF gpt-oss support and Qwen3.5 chat streaming failure are unrelated." + }, + { + "left": "issue:45020", + "right": "issue:45800", + "accept": false, + "reason": "Remote_code model breakage and a torch version incompatibility are too broad and distinct to treat as one bug." + }, + { + "left": "issue:43531", + "right": "issue:45070", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and a pydantic field regression are different issues." + }, + { + "left": "issue:43541", + "right": "issue:44589", + "accept": false, + "reason": "Grouped_mm tracing failure and missing Float8 storage object are unrelated runtime errors." + }, + { + "left": "issue:41093", + "right": "issue:43596", + "accept": false, + "reason": "Both are index errors, but one is a mask-length mismatch and the other is a zero-size tensor access during ZeRO-3 init." + }, + { + "left": "issue:43054", + "right": "issue:43994", + "accept": true, + "reason": "Both point to Siglip2 producing bad/nonsensical text-embedding-driven results, likely the same underlying model/processor issue." + }, + { + "left": "issue:45542", + "right": "issue:45800", + "accept": false, + "reason": "TensorBoard-without-TensorFlow backend error and torch/transformers incompatibility are unrelated." + }, + { + "left": "issue:45759", + "right": "issue:45902", + "accept": false, + "reason": "Composite Qwen config unwrapping and uninitialized rotary embedding memory are different initialization bugs." + }, + { + "left": "issue:39186", + "right": "issue:44918", + "accept": false, + "reason": "FSDP 2-D weight assertion and TRL embedding unpacking failure are different problems." + }, + { + "left": "issue:43716", + "right": "issue:44918", + "accept": false, + "reason": "Mistral-3 image-preprocessor dtype mismatch and Qwen3.5 TRL unpacking failure are unrelated." + }, + { + "left": "issue:40990", + "right": "issue:44960", + "accept": false, + "reason": "GPT-OSS perplexity quality issue and GLM5 are not the same bug." + }, + { + "left": "issue:45632", + "right": "issue:45698", + "accept": false, + "reason": "Both involve custom-module loading, but one is a trust_remote_code cache path collision and the other is a wrong module loaded after save_pretrained; not enough evidence of the same concrete bug." + }, + { + "left": "issue:43645", + "right": "issue:45399", + "accept": false, + "reason": "Jupyter custom-model initialization regression and flash-attn2 fallback blocking are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45005", + "accept": false, + "reason": "Mistral3 lm_head tying and translation-model weight tying are similar in theme but not the same concrete bug." + }, + { + "left": "issue:44530", + "right": "issue:45005", + "accept": false, + "reason": "PagedAttention linear_attention crash and translation-model tied weights are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is conservative and stays grounded in the packet: most pairs are plausibly distinct based on titles, and the single accepted Siglip2 pair is the only one with a reasonable shared-bug reading. The choice of issue:45020 as a representative is subjective but defensible given its broad compatibility/regression framing." + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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: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:43873|issue:45005", + "issue:44360|issue:44512", + "issue:44863|issue:45005", + "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:43872|issue:44589", + "issue:43950|issue:45902", + "issue:43866|issue:44403", + "issue:43054|issue:45701", + "issue:41553|issue:43883", + "issue:38175|issue:43054", + "issue:44910|issue:45859", + "issue:45663|issue:45902", + "issue:41553|issue:43881", + "issue:42947|issue:44387", + "issue:43054|issue:43493", + "issue:44403|issue:44451", + "issue:43643|issue:45020", + "issue:43975|issue:45237", + "issue:44403|issue:44488", + "issue:42947|issue:43856", + "issue:41762|issue:42915", + "issue:45685|issue:45823" + ], + "prepared_review_unit_hash": "10c469a24391574294c85ab89264bc5068c59504979715b6ecb21e8fd2154ca9", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36380, + "estimated_input_tokens": 9095, + "estimated_eval_tokens": 18446 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a mixed similarity cluster. The safest duplicate chain is the v5 non-persistent-buffer regression around 43950 and 44534; 45902 is related but not strong enough to confirm from the packet alone. The SigLIP2 pair 38175/43054 also looks related, but the evidence is too thin to call it a confirmed duplicate. The Qwen3.5 3D position_ids pair (44910/44928) is the other clear duplicate-like edge.", + "confidence": 0.72, + "canonical_issue_reason": "43950 is still the clearest root-cause report for the non-persistent-buffer v5 regression, with 44534 as the strongest corroborating duplicate-like report.", + "canonical_pr_reason": null, + "best_issue_reason": "43950 is the best representative issue because it states the regression most concretely and anchors the strongest duplicate chain in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42915", + "right": "issue:45399", + "accept": false, + "reason": "Different failures: FP8/Qwen3 MoE loading vs flash-attn fallback gating; same area, not the same bug." + }, + { + "left": "issue:43873", + "right": "issue:45005", + "accept": false, + "reason": "Quantization/offloading memory behavior vs tied-weights translation-model bug; no shared concrete code path." + }, + { + "left": "issue:44360", + "right": "issue:44512", + "accept": false, + "reason": "DSA indexer math issue vs a docs typo; unrelated." + }, + { + "left": "issue:44863", + "right": "issue:45005", + "accept": false, + "reason": "NemotronH checkpoint-loading failure vs translation tied-weights regression; different model paths." + }, + { + "left": "issue:44918", + "right": "issue:45507", + "accept": false, + "reason": "Qwen3.5 embedding unpacking in TRL vs GraniteMoE invalid method call; not the same defect." + }, + { + "left": "issue:44492", + "right": "issue:44512", + "accept": false, + "reason": "Cache-strategy typo vs removed-command documentation; editorial, not duplicate bugs." + }, + { + "left": "issue:43975", + "right": "issue:44292", + "accept": false, + "reason": "Detokenization regression vs Qwen-3 NVFP4 load/run failure; different symptoms and paths." + }, + { + "left": "issue:44292", + "right": "issue:45507", + "accept": false, + "reason": "Separate model-loading/runtime failures with no clear shared root cause." + }, + { + "left": "issue:43950", + "right": "issue:44534", + "accept": true, + "reason": "Same non-persistent-buffer v5 regression; 44534 matches the junk/corruption symptom reported in 43950." + }, + { + "left": "issue:43643", + "right": "issue:43950", + "accept": false, + "reason": "Remote-code config-field loss vs non-persistent-buffer corruption; different bugs." + }, + { + "left": "issue:39692", + "right": "issue:43054", + "accept": false, + "reason": "Docs example/quantization failure vs SigLIP2 embedding-quality regression; too different." + }, + { + "left": "issue:43828", + "right": "issue:44928", + "accept": false, + "reason": "Autocast dtype mismatch on Phi-tiny-MoE vs Qwen3.5 3D position_ids/SDPA instability; not the same issue." + }, + { + "left": "issue:36331", + "right": "issue:41762", + "accept": false, + "reason": "CustomTrainer API change vs Gemma3 ZeRO-3 loading crash; unrelated." + }, + { + "left": "issue:41628", + "right": "issue:45800", + "accept": false, + "reason": "Import/export issue for AutoImageProcessor vs torch/transformers version incompatibility; not a duplicate." + }, + { + "left": "issue:43872", + "right": "issue:44589", + "accept": false, + "reason": "bitsandbytes kwarg incompatibility vs missing Float8 storage; different incompatibilities." + }, + { + "left": "issue:43950", + "right": "issue:45902", + "accept": false, + "reason": "Related non-persistent-buffer regression, but the packet does not firmly establish the same concrete bug/root cause; safer as a candidate duplicate than a confirmed one." + }, + { + "left": "issue:43866", + "right": "issue:44403", + "accept": false, + "reason": "Checkpoint corruption vs generic loading noise; no shared defect." + }, + { + "left": "issue:43054", + "right": "issue:45701", + "accept": false, + "reason": "SigLIP2 embedding-quality regression vs tokenizer/version behavior change; different subsystems." + }, + { + "left": "issue:41553", + "right": "issue:43883", + "accept": false, + "reason": "Bad tokenizer error message vs missing tied-weights attribute; unrelated." + }, + { + "left": "issue:38175", + "right": "issue:43054", + "accept": false, + "reason": "Both mention SigLIP2 output quality problems, but the packet does not clearly show the same bug/root cause; too weak to confirm as duplicates." + }, + { + "left": "issue:44910", + "right": "issue:45859", + "accept": false, + "reason": "Qwen3.5 flash-attn/packed-sequence crash vs CPU-offload incompatibility in rotary embedding; different code paths." + }, + { + "left": "issue:45663", + "right": "issue:45902", + "accept": false, + "reason": "Gemma4 FSDP2 KeyError vs uninitialized rotary inv_freq after to_empty(); different models and failures." + }, + { + "left": "issue:41553", + "right": "issue:43881", + "accept": false, + "reason": "Voxtral tokenizer error message vs glm-4v loading failure; unrelated model-loading problems." + }, + { + "left": "issue:42947", + "right": "issue:44387", + "accept": false, + "reason": "Gradient-checkpointing/LoRA issue vs int4 quantization reserved-memory OOM; different root causes." + }, + { + "left": "issue:43054", + "right": "issue:43493", + "accept": false, + "reason": "Very related SigLIP2 reports, but one is a broad HF-vs-JAX discrepancy and the other is a symptom report; not enough to call them the same issue." + }, + { + "left": "issue:44403", + "right": "issue:44451", + "accept": false, + "reason": "General loading noise vs a specific model load failure; not the same bug." + }, + { + "left": "issue:43643", + "right": "issue:45020", + "accept": false, + "reason": "Specific AutoConfig remote_code field loss vs broad remote_code breakage report; too generic to merge confidently." + }, + { + "left": "issue:43975", + "right": "issue:45237", + "accept": false, + "reason": "Detokenization bug vs AMD GPU support/loading problem; unrelated." + }, + { + "left": "issue:44403", + "right": "issue:44488", + "accept": false, + "reason": "Generic loading noise vs a specific model load failure; distinct reports." + }, + { + "left": "issue:42947", + "right": "issue:43856", + "accept": false, + "reason": "LoRA checkpointing issue vs Qwen3 MoE memory usage concern; different problems." + }, + { + "left": "issue:41762", + "right": "issue:42915", + "accept": false, + "reason": "Gemma3 ZeRO-3 loading crash vs Qwen3 MoE FP8/flash-attn loading issue; unrelated." + }, + { + "left": "issue:45685", + "right": "issue:45823", + "accept": false, + "reason": "MPS histogram op missing in MoE vs Gemma4 device-mismatch forward bug; different defects." + }, + { + "left": "issue:44910", + "right": "issue:44928", + "accept": true, + "reason": "Same Qwen3.5 3D position_ids handling bug; one crashes flash-attn, the other triggers SDPA/BF16 instability." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is mostly grounded, but a couple of the accepted soft-edge pairs are a bit too strong for the evidence in the packet. In particular, 43950\u219445902 and 38175\u219443054 look related, but the titles alone do not clearly establish the same bug/root cause; they are safer as candidate duplicates than confirmed ones. Because the task asks for conservative soft-edge verdicts, I\u2019d reject this as slightly overconfident." + }, + "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": 4, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:24643", + "issue:30064", + "issue:36331", + "issue:41093", + "issue:41553", + "issue:42491", + "issue:42915", + "issue:42947", + "issue:43257", + "issue:43404", + "issue:43638", + "issue:43742", + "issue:43782", + "issue:43792", + "issue:43824", + "issue:43827", + "issue:43828", + "issue:43854", + "issue:43866", + "issue:43872", + "issue:43881", + "issue:43883", + "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:43881|issue:43883", + "issue:43824|issue:43872", + "issue:44991|issue:45399", + "issue:45020|issue:45823", + "issue:42491|issue:44368", + "issue:44661|issue:45020", + "issue:43828|issue:45823", + "issue:45081|issue:45399", + "issue:45020|issue:45356", + "issue:43827|issue:44509", + "issue:43742|issue:45399", + "issue:43404|issue:43950", + "issue:42915|issue:45020", + "issue:43638|issue:45823", + "issue:43257|issue:43866", + "issue:43792|issue:44220", + "issue:43872|issue:44291", + "issue:42915|issue:43872", + "issue:41553|issue:44991", + "issue:43940|issue:45313", + "issue:36331|issue:41093", + "issue:42947|issue:45923", + "issue:44928|issue:45588", + "issue:43854|issue:44863", + "issue:43257|issue:45305", + "issue:43872|issue:45538", + "issue:41093|issue:44805", + "issue:44479|issue:44560", + "issue:41553|issue:44488", + "issue:44530|issue:44863", + "issue:24643|issue:30064" + ], + "prepared_review_unit_hash": "4be4c4b508b5552986984d17dad15c6780305489344efa22e2b1b7abdaacb436", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37611, + "estimated_input_tokens": 9403, + "estimated_eval_tokens": 19062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly heterogeneous cluster. The clearest duplicate pairs are the v5 docs pipeline-removal reports and the init_empty_weights/_is_hf_initialized loading regression; there is also a plausible Qwen3-VL video-input regression pair, but the rest look unrelated by model, subsystem, or failure mode.", + "confidence": 0.83, + "canonical_issue_reason": "No single cluster-wide canonical stands out, but if one issue must anchor triage, 44291 is the strongest because it states the exact root cause/version boundary and matches the clearest duplicate counterpart (43872).", + "canonical_pr_reason": null, + "best_issue_reason": "44291 is the most precise, actionable issue in the set and best represents the concrete loading-regression duplicate pair; the other items are too mixed to serve as a better global anchor.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43782", + "right": "issue:44960", + "accept": false, + "reason": "Different models and failures: Qwen3-VL weight_only loading vs a GLM5 unit-test loading problem." + }, + { + "left": "issue:43881", + "right": "issue:43883", + "accept": false, + "reason": "Different code paths: GLM-4V loading failure vs missing all_tied_weights_keys on Molmo." + }, + { + "left": "issue:43824", + "right": "issue:43872", + "accept": false, + "reason": "ImportError for a missing Qwen2_5_VL class vs a bitsandbytes init TypeError; unrelated." + }, + { + "left": "issue:44991", + "right": "issue:45399", + "accept": false, + "reason": "Tokenizer loading regression vs flash-attn fallback gating; not the same bug." + }, + { + "left": "issue:45020", + "right": "issue:45823", + "accept": false, + "reason": "Generic remote_code breakage vs Gemma4 device-map forward mismatch; different code paths." + }, + { + "left": "issue:42491", + "right": "issue:44368", + "accept": false, + "reason": "Both mention Qwen/LoRA, but one is a model-compatibility regression and the other is just a tie_word_embeddings warning." + }, + { + "left": "issue:44661", + "right": "issue:45020", + "accept": false, + "reason": "add-new-model-like handling of TOKENIZER_MAPPING_NAMES is a different failure from remote_code model loading regressions." + }, + { + "left": "issue:43828", + "right": "issue:45823", + "accept": false, + "reason": "Autocast dtype mismatch and device placement mismatch are distinct runtime problems." + }, + { + "left": "issue:45081", + "right": "issue:45399", + "accept": false, + "reason": "Mistral regex patch crash vs flash-attn fallback selection; unrelated subsystems." + }, + { + "left": "issue:45020", + "right": "issue:45356", + "accept": false, + "reason": "Remote_code loading regressions vs Kimi tokenizer codec handling; different underlying bugs." + }, + { + "left": "issue:43827", + "right": "issue:44509", + "accept": true, + "reason": "Same docs regression: v5 removed pipeline tasks, but the docs still reference summarization/translation/text2text-generation pipeline usage." + }, + { + "left": "issue:43742", + "right": "issue:45399", + "accept": false, + "reason": "MobileLLM key error vs flash-attn installation/fallback logic; not the same issue." + }, + { + "left": "issue:43404", + "right": "issue:43950", + "accept": false, + "reason": "Untied lm_head in Mistral3 and non-persistent buffer corruption are different model/runtime bugs." + }, + { + "left": "issue:42915", + "right": "issue:45020", + "accept": false, + "reason": "Qwen3 MOE weight conversion under accelerate+deepspeed is a different concrete failure than generic remote_code breakage." + }, + { + "left": "issue:43638", + "right": "issue:45823", + "accept": false, + "reason": "ZeRO-3 index-out-of-bounds on a non-pretrained Bert model vs Gemma4 PLE device mismatch; unrelated." + }, + { + "left": "issue:43257", + "right": "issue:43866", + "accept": false, + "reason": "Qwen3 MOE weight conversion bug vs Ovis2 checkpoint corruption; different models and symptoms." + }, + { + "left": "issue:43792", + "right": "issue:44220", + "accept": false, + "reason": "Whisper startup failure and _torch_extract_fbank_features are related at a very high level, but the titles do not show the same concrete bug or code path." + }, + { + "left": "issue:43872", + "right": "issue:44291", + "accept": true, + "reason": "Same init_empty_weights / _is_hf_initialized incompatibility; 44291 spells out the exact regression that 43872 reports more generally." + }, + { + "left": "issue:42915", + "right": "issue:43872", + "accept": false, + "reason": "DeepSpeed Qwen3 MOE weight-loading bug vs bitsandbytes init error; different failure modes." + }, + { + "left": "issue:41553", + "right": "issue:44991", + "accept": false, + "reason": "Bad AutoTokenizer error messaging for Voxtral vs tokenizer loading failure for EMBEDDIA/est-roberta." + }, + { + "left": "issue:43940", + "right": "issue:45313", + "accept": false, + "reason": "Both are ZeRO-3 weight-loading reports, but they target different Qwen variants and the titles do not indicate the same concrete root cause." + }, + { + "left": "issue:36331", + "right": "issue:41093", + "accept": false, + "reason": "Trainer compute_loss API mismatch vs a mask/tensor shape IndexError; unrelated." + }, + { + "left": "issue:42947", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing ineffective with PEFT LoRA and missing supports_gradient_checkpointing are related broadly, but not the same bug." + }, + { + "left": "issue:44928", + "right": "issue:45588", + "accept": false, + "reason": "RLHF NaNs from 3D position_ids/SDPA fallback vs flash_attention crash on s_aux=None; different mechanisms." + }, + { + "left": "issue:43854", + "right": "issue:44863", + "accept": false, + "reason": "GLM-4.7 loading in tests vs NemotronH checkpoint loading; unrelated model-specific issues." + }, + { + "left": "issue:43257", + "right": "issue:45305", + "accept": false, + "reason": "MOE weight conversion during loading vs gradient averaging under GAS with DeepSpeed; different training bugs." + }, + { + "left": "issue:43872", + "right": "issue:45538", + "accept": false, + "reason": "bitsandbytes init incompatibility vs CLIPTokenizer max_length default; unrelated." + }, + { + "left": "issue:41093", + "right": "issue:44805", + "accept": false, + "reason": "The error shape is similar, but the titles do not establish the same root cause or model path; likely separate mask-shape bugs." + }, + { + "left": "issue:44479", + "right": "issue:44560", + "accept": true, + "reason": "Both point to the Qwen3-VL video-input regression in 5.3.0; 44560 looks like the embedding-model manifestation of the same StopIteration failure." + }, + { + "left": "issue:41553", + "right": "issue:44488", + "accept": false, + "reason": "Voxtral tokenizer error messaging vs sleng-bert loading failure; different artifacts and failures." + }, + { + "left": "issue:44530", + "right": "issue:44863", + "accept": false, + "reason": "PagedAttentionCache linear_attention crash vs NemotronH checkpoint loading; different subsystems and bugs." + }, + { + "left": "issue:24643", + "right": "issue:30064", + "accept": false, + "reason": "DeepSpeed training weight shape error vs image processor void-segmentation-map failure; unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The accepted soft-edge pairs are plausibly the same underlying bug family, and the rejected pairs are generally justified by distinct model/subsystem/failure-mode differences. No overstatement stands out." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 5, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43828", + "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:45313", + "issue:45325", + "issue:45357", + "issue:45584", + "issue:45663", + "issue:45685", + "issue:45823", + "issue:45834", + "issue:45902" + ], + "soft_pairs": [ + "issue:43366|issue:43828", + "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:42915|issue:43957", + "issue:43366|issue:44589", + "issue:43645|issue:43646", + "issue:44560|issue:44918", + "issue:42617|issue:43366", + "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:45313|issue:45357", + "issue:42915|issue:44756" + ], + "prepared_review_unit_hash": "e6bfe591091595960c9618c14527c1547186951444b0889dc1e2fd75f6668dac", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36659, + "estimated_input_tokens": 9165, + "estimated_eval_tokens": 18586 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous, but a few tight duplicate subgroups stand out: non-persistent buffer regressions, SigLIP2 output corruption, Qwen3.5 flash-attention failures, and a couple of model-loading regressions.", + "confidence": 0.76, + "canonical_issue_reason": "issue:43644 is the clearest and earliest report of the exact non-persistent-buffer regression, and it has a near-exact duplicate in issue:44534.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43644 is the best overall issue anchor here: specific, regression-focused, and the strongest exact-match duplicate candidate in the set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43366", + "right": "issue:43828", + "accept": false, + "reason": "Different models and failure modes; no shared concrete code path." + }, + { + "left": "issue:45325", + "right": "issue:45902", + "accept": false, + "reason": "Different bug classes: rope-position scaling vs uninitialized rotary embedding memory." + }, + { + "left": "issue:45663", + "right": "issue:45823", + "accept": false, + "reason": "Same model family, but different failures in different paths (FSDP training KeyError vs device-map forward mismatch)." + }, + { + "left": "issue:43366", + "right": "issue:43531", + "accept": false, + "reason": "Unrelated Qwen3-MoE sliding-window bug vs GGUF gpt-oss support request." + }, + { + "left": "issue:44155", + "right": "issue:45584", + "accept": false, + "reason": "Different audio/ASR paths and distinct symptoms; not the same underlying bug." + }, + { + "left": "issue:43950", + "right": "issue:45834", + "accept": false, + "reason": "Non-persistent-buffer regression vs long-OCR index error; unrelated code paths." + }, + { + "left": "issue:44560", + "right": "issue:45685", + "accept": false, + "reason": "Different subsystems and errors; no plausible single fix for both." + }, + { + "left": "issue:43824", + "right": "issue:43957", + "accept": false, + "reason": "Import surface issue vs meta-device loading bug; different concrete problems." + }, + { + "left": "issue:43550", + "right": "issue:44910", + "accept": false, + "reason": "Separate attention-path bugs in different models; not mergeable into one fix." + }, + { + "left": "issue:42915", + "right": "issue:43957", + "accept": false, + "reason": "FP8/Qwen3-MoE failure is unrelated to meta-device loading." + }, + { + "left": "issue:43366", + "right": "issue:44589", + "accept": false, + "reason": "GGUF support request vs float8 storage lookup error; unrelated issues." + }, + { + "left": "issue:43645", + "right": "issue:43646", + "accept": true, + "reason": "Both describe the same Transformers v5 custom-model initialization regression; one is just the notebook-specific report." + }, + { + "left": "issue:44560", + "right": "issue:44918", + "accept": false, + "reason": "Video StopIteration and SFT embedding unpacking are different Qwen3.5 failures." + }, + { + "left": "issue:42617", + "right": "issue:43366", + "accept": false, + "reason": "No shared bug; one is a script/runtime failure, the other is GGUF architecture support." + }, + { + "left": "issue:36010", + "right": "issue:42915", + "accept": false, + "reason": "ImportError for GenerationMixin is unrelated to Qwen3-MoE FP8 behavior." + }, + { + "left": "issue:44661", + "right": "issue:45823", + "accept": false, + "reason": "Tokenizer-mapping model discovery bug vs Gemma4 PLE device mismatch; different code paths." + }, + { + "left": "issue:44560", + "right": "issue:44805", + "accept": false, + "reason": "Different error types and model-specific failures." + }, + { + "left": "issue:43632", + "right": "issue:45834", + "accept": false, + "reason": "_is_hf_initialized regression is unrelated to Kosmos2.5 long-input indexing." + }, + { + "left": "issue:38175", + "right": "issue:43994", + "accept": true, + "reason": "Same SigLIP2 model and same wrong-output symptom family; these look like duplicate reports of one inference bug." + }, + { + "left": "issue:43638", + "right": "issue:44661", + "accept": false, + "reason": "DeepSpeed ZeRO-3 Bert indexing failure is unrelated to tokenizer mapping-name handling." + }, + { + "left": "issue:43596", + "right": "issue:44661", + "accept": false, + "reason": "Same as above: ZeRO-3/Bert init failure vs tokenizer mapping bug." + }, + { + "left": "issue:44661", + "right": "issue:44805", + "accept": false, + "reason": "Tokenizer-mapping issue and mask-shape IndexError are not the same bug." + }, + { + "left": "issue:44877", + "right": "issue:45030", + "accept": false, + "reason": "Both involve config validation, but the titles point to different models and distinct failing checks." + }, + { + "left": "issue:43317", + "right": "issue:44756", + "accept": false, + "reason": "Device-map/offload loading failure is unrelated to mmap OOM on Strix Halo." + }, + { + "left": "issue:43644", + "right": "issue:44534", + "accept": true, + "reason": "Exact same non-persistent-buffer junk regression, just reported with different wording." + }, + { + "left": "issue:38175", + "right": "issue:39692", + "accept": false, + "reason": "Both are SigLIP2-related, but one is bad model outputs and the other is a docs/example mismatch plus quantization failure." + }, + { + "left": "issue:42491", + "right": "issue:43720", + "accept": false, + "reason": "LoRA/qwen3_moe version compatibility and BitNet packed-weight loading are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:42915", + "accept": false, + "reason": "Different Qwen3 failures: device-map CUDA assert vs FineGrainedFP8Config." + }, + { + "left": "issue:44960", + "right": "issue:44977", + "accept": false, + "reason": "Different GLM/Qwen3.5 issues with different failure modes." + }, + { + "left": "issue:44451", + "right": "issue:44488", + "accept": true, + "reason": "Both are reports that current/latest Transformers cannot load specific BERT-family models; likely the same loader regression." + }, + { + "left": "issue:45313", + "right": "issue:45357", + "accept": false, + "reason": "Different Qwen3.5 issues affecting separate components (weight loading vs save_pretrained key handling)." + }, + { + "left": "issue:42915", + "right": "issue:44756", + "accept": false, + "reason": "FP8 Qwen3-MoE failure and Strix Halo mmap OOM are unrelated." + }, + { + "left": "issue:44910", + "right": "issue:44977", + "accept": true, + "reason": "Both are Qwen3.5 flash-attention generation failures and likely the same underlying attention/position-id bug." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the accepted soft edges are reasonably conservative given the titles and descriptions. The canonical issue choice for the non-persistent-buffer regression is well supported by the near-exact duplicate pair, and the other accepted groups are plausible duplicate/near-duplicate bug reports rather than clearly separate failures." + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:36010", + "issue:39404", + "issue:39692", + "issue:41720", + "issue:42491", + "issue:42947", + "issue:43257", + "issue:43335", + "issue:43404", + "issue:43425", + "issue:43454", + "issue:43644", + "issue:43645", + "issue: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:45216", + "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:42491|issue:45276", + "issue:43854|issue:43931", + "issue:45488|issue:45812", + "issue:43257|issue:43854", + "issue:44488|issue:44991", + "issue:39692|issue:43994", + "issue:43644|issue:43645", + "issue:43856|issue:43873", + "issue:42491|issue:43931", + "issue:45362|issue:45406", + "issue:43856|issue:44387", + "issue:44960|issue:46050", + "issue:42491|issue:45292", + "issue:44910|issue:45588", + "issue:44164|issue:44291", + "issue:36010|issue:39404", + "issue:45276|issue:45335", + "issue:43425|issue:43957", + "issue:43404|issue:43454", + "issue:41720|issue:44155", + "issue:44402|issue:44488", + "issue:39692|issue:44960", + "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:43854|issue:45216", + "issue:45276|issue:45292", + "issue:44164|issue:44756" + ], + "prepared_review_unit_hash": "123b51e1b86a1ede2c73727e715132f9102ce9430e3d037b65eced5ebeb6dbc1", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 36, + "serialized_chars": 38645, + "estimated_input_tokens": 9662, + "estimated_eval_tokens": 19580 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most items are unrelated single-issue reports; the only clear duplicate family is around `resize_token_embeddings` not propagating resized embeddings/output heads across model variants. I would keep issue 45276 as the representative for that family.", + "confidence": 0.72, + "canonical_issue_reason": "Issue 45276 is the most specific and central report in the resize-token-embeddings family: it names the failing API and covers both input and output embedding propagation, which subsumes the more generic 45292 and the model-specific 45335 symptoms.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45276 is the best global issue representative because it is detailed enough to anchor the underlying bug while still broad enough to represent the closely related resize-token-embeddings reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43708", + "right": "issue:44075", + "accept": false, + "reason": "Different bugs: checkpoint resume step accounting vs SGD optimizer arguments not being applied." + }, + { + "left": "issue:42491", + "right": "issue:45276", + "accept": false, + "reason": "Both involve model usage/regression, but the concrete failures are unrelated: Qwen3 MoE LoRA compatibility vs token embedding resizing." + }, + { + "left": "issue:43854", + "right": "issue:43931", + "accept": false, + "reason": "Both mention Qwen3-VL loading, but one is a generic unit-test load failure and the other is a specific shape-mismatch error; not clearly the same bug." + }, + { + "left": "issue:45488", + "right": "issue:45812", + "accept": false, + "reason": "Same broad tokenizer area, but different model families and different failure modes; too weak to treat as the same underlying bug." + }, + { + "left": "issue:43257", + "right": "issue:43854", + "accept": false, + "reason": "Different problems: MoE weight conversion under accelerate+deepspeed vs model loading/test issues for Qwen3-VL." + }, + { + "left": "issue:44488", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-loading regressions, but they affect different models and likely different code paths." + }, + { + "left": "issue:39692", + "right": "issue:43994", + "accept": false, + "reason": "SigLIP2 doc/example errors and nonsensical runtime outputs are related to the same model family, but the reported bugs are not the same concrete issue." + }, + { + "left": "issue:43644", + "right": "issue:43645", + "accept": false, + "reason": "Separate 5.0 regressions: junk-filled non-persistent buffers vs Jupyter custom model initialization." + }, + { + "left": "issue:43856", + "right": "issue:43873", + "accept": false, + "reason": "Different quantization/offloading/memory complaints; not the same underlying bug." + }, + { + "left": "issue:42491", + "right": "issue:43931", + "accept": false, + "reason": "Different concrete failures: LoRA compatibility across versions vs weight-shape mismatch while loading a VL model." + }, + { + "left": "issue:45362", + "right": "issue:45406", + "accept": false, + "reason": "Both are chat/processor-related model issues, but one is Qwen3.5 chat crashes and the other is a Gemma4 processor AttributeError." + }, + { + "left": "issue:43856", + "right": "issue:44387", + "accept": false, + "reason": "Both concern memory, but one is Qwen3 MoE training efficiency and the other is int4 quantization reserved-memory/OOM regression." + }, + { + "left": "issue:44960", + "right": "issue:46050", + "accept": false, + "reason": "Different model/version topics: GLM5 versus a sinq save/load corruption report." + }, + { + "left": "issue:42491", + "right": "issue:45292", + "accept": false, + "reason": "Unrelated issues: Qwen3 MoE LoRA compatibility vs embedding resize propagation." + }, + { + "left": "issue:44910", + "right": "issue:45588", + "accept": false, + "reason": "Both touch flash attention, but the failures are distinct code paths: packed-sequence misinterpretation vs `s_aux=None` AttributeError." + }, + { + "left": "issue:44164", + "right": "issue:44291", + "accept": false, + "reason": "Both involve model loading/saving internals, but extra_state handling and `_is_hf_initialized` argument handling are different bugs." + }, + { + "left": "issue:36010", + "right": "issue:39404", + "accept": false, + "reason": "ImportError for `GenerationMixin` is unrelated to Whisper `return_language` pipeline behavior." + }, + { + "left": "issue:45276", + "right": "issue:45335", + "accept": true, + "reason": "Both report the same underlying `resize_token_embeddings` propagation bug: resized embeddings are not reflected in model-specific output/input embedding modules." + }, + { + "left": "issue:43425", + "right": "issue:43957", + "accept": false, + "reason": "Torch compatibility and meta-device loading failures are different regressions." + }, + { + "left": "issue:43404", + "right": "issue:43454", + "accept": false, + "reason": "Both are weight-tying issues, but for different models and different symptoms; not the same concrete bug." + }, + { + "left": "issue:41720", + "right": "issue:44155", + "accept": false, + "reason": "Qwen3 auto device-map CUDA assert and AudioFlamingo3 batched inference token leakage are unrelated." + }, + { + "left": "issue:44402", + "right": "issue:44488", + "accept": false, + "reason": "Tokenizer vocab-size mismatch for one model is not the same as a separate model-loading failure for another." + }, + { + "left": "issue:39692", + "right": "issue:44960", + "accept": false, + "reason": "SigLIP2 documentation/model issues are unrelated to the GLM5 report." + }, + { + "left": "issue:42947", + "right": "issue:45704", + "accept": false, + "reason": "Gradient checkpointing with PEFT LoRA and an apex RMSNorm memory leak are different bugs." + }, + { + "left": "issue:45663", + "right": "issue:45759", + "accept": false, + "reason": "Gemma4 FSDP2 KeyError and Qwen3.5 multimodal config unwrapping are different failure modes." + }, + { + "left": "issue:43873", + "right": "issue:44387", + "accept": false, + "reason": "Quantization/offloading issue vs int4 reserved-memory regression; same area, different bugs." + }, + { + "left": "issue:44291", + "right": "issue:44756", + "accept": false, + "reason": "Meta-init loading argument mismatch and mmap-related OOM are unrelated." + }, + { + "left": "issue:45663", + "right": "issue:45859", + "accept": false, + "reason": "Gemma4 FSDP2 attention state handling and Qwen3_5Moe CPU-offload incompatibility are distinct." + }, + { + "left": "issue:44155", + "right": "issue:45823", + "accept": false, + "reason": "AudioFlamingo3 batched inference leakage and Gemma4 PLE device mismatch are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:45663", + "accept": false, + "reason": "Sparse-layer creation in SwitchTransformers and Gemma4 FSDP2 shared_kv_state rebuilding are unrelated." + }, + { + "left": "issue:44292", + "right": "issue:44912", + "accept": false, + "reason": "Different quantization/model-loading problems for different models." + }, + { + "left": "issue:44387", + "right": "issue:44756", + "accept": false, + "reason": "Different causes of OOM: int4 reserved memory growth vs mmap on Strix Halo." + }, + { + "left": "issue:43957", + "right": "issue:44756", + "accept": false, + "reason": "Meta-device loading breakage and mmap/OOM are different issues." + }, + { + "left": "issue:43854", + "right": "issue:45216", + "accept": false, + "reason": "Unit-test model loading failure for GLM-4.7-Flash vs Qwen3.5 save_pretrained regression are different bugs." + }, + { + "left": "issue:45276", + "right": "issue:45292", + "accept": true, + "reason": "Same underlying `resize_token_embeddings` bug: resized embeddings are not reflected in output embeddings." + }, + { + "left": "issue:44164", + "right": "issue:44756", + "accept": false, + "reason": "Extra_state serialization and Strix Halo mmap OOM are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: 45276, 45292, and 45335 clearly form the resize_token_embeddings family, and choosing 45276 as the representative is conservative and well supported by the titles. The soft-edge verdicts are also generally cautious, mostly rejecting pairs with different model families or failure modes rather than overclaiming duplicates. I don\u2019t see an overstatement that would justify rejection." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 7, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:39404", + "issue:42947", + "issue:43257", + "issue:43299", + "issue:43381", + "issue:43425", + "issue:43596", + "issue:43611", + "issue:43645", + "issue:43646", + "issue:43708", + "issue:43716", + "issue:43828", + "issue:43901", + "issue:43906", + "issue:43931", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44164", + "issue:44265", + "issue:44402", + "issue:44451", + "issue:44479", + "issue:44509", + "issue:44530", + "issue:44589", + "issue:44671", + "issue:44683", + "issue:44756", + "issue:44805", + "issue:44863", + "issue:44877", + "issue:44898", + "issue:44912", + "issue:44918", + "issue:44928", + "issue:45084", + "issue:45292", + "issue:45335", + "issue:45588", + "issue:45663", + "issue:45684", + "issue:45698", + "issue:45701", + "issue:45759", + "issue:45823" + ], + "soft_pairs": [ + "issue:43381|issue:44928", + "issue:45684|issue:45698", + "issue:42947|issue:43381", + "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:43257|issue:44863", + "issue:44589|issue:44756", + "issue:44683|issue:45823", + "issue:39404|issue:45701", + "issue:43425|issue:45823", + "issue:43425|issue:43716", + "issue:43611|issue:43645", + "issue:43906|issue:44530", + "issue:44530|issue:45084", + "issue:39404|issue:44912", + "issue:43901|issue:44509", + "issue:44265|issue:44898", + "issue:44479|issue:44805", + "issue:43994|issue:44079", + "issue:43425|issue:44918", + "issue:43299|issue:43931", + "issue:44589|issue:45084", + "issue:43596|issue:44683", + "issue:39404|issue:44671", + "issue:44164|issue:45759" + ], + "prepared_review_unit_hash": "377f2324901e6c78faabdbeffda8ff3d57607ee138bc54e61d0ab6f549d2a618", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 31, + "serialized_chars": 36466, + "estimated_input_tokens": 9117, + "estimated_eval_tokens": 18490 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a grab-bag of unrelated Transformers regressions; only the resize_token_embeddings pair looks like a true duplicate. Most other pairs share a subsystem or vague symptom but not the same underlying bug.", + "confidence": 0.88, + "canonical_issue_reason": "issue:45292 is the broadest and most central formulation of the resize_token_embeddings bug; issue:45335 reads like a model-specific manifestation of the same problem.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45292 is the best canonical issue because it describes the generic behavior regression and subsumes the more specific t5gemma report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43381", + "right": "issue:44928", + "accept": false, + "reason": "Both involve gradient-related training behavior, but one is an eval-mode checkpointing restriction and the other is a Qwen3.5 RLHF overflow/SDPA fallback bug." + }, + { + "left": "issue:45684", + "right": "issue:45698", + "accept": false, + "reason": "Both mention save_pretrained/from_pretrained, but one is about read-only file permissions during save and the other is about loading the wrong custom module." + }, + { + "left": "issue:42947", + "right": "issue:43381", + "accept": false, + "reason": "LoRA gradient checkpointing ineffective is a different bug from gradient checkpointing being disallowed in eval mode." + }, + { + "left": "issue:44479", + "right": "issue:44877", + "accept": false, + "reason": "Different model families and failure modes: video-input regression vs strict config blocking granite_speech loading." + }, + { + "left": "issue:45588", + "right": "issue:45663", + "accept": false, + "reason": "A flash_attention crash on sink-less models is not the same as a Gemma4 FSDP2 shared_kv_states KeyError." + }, + { + "left": "issue:43646", + "right": "issue:43708", + "accept": false, + "reason": "Custom model initialization in notebooks is unrelated to resume_from_checkpoint max_steps calculation." + }, + { + "left": "issue:43645", + "right": "issue:44164", + "accept": false, + "reason": "Notebook-defined custom model initialization and extra_state save/load handling are different code paths and bugs." + }, + { + "left": "issue:43828", + "right": "issue:43957", + "accept": false, + "reason": "Both are dtype/device-related, but one is an autocast mismatch for Phi-tiny-MoE and the other is a meta-device loading regression." + }, + { + "left": "issue:45292", + "right": "issue:45335", + "accept": true, + "reason": "These describe the same resize_token_embeddings regression; 45335 is a model-specific instance of the broader output_embeddings issue." + }, + { + "left": "issue:44402", + "right": "issue:44451", + "accept": false, + "reason": "Both concern loading/tokenizer oddities, but the affected models and likely root causes are different." + }, + { + "left": "issue:43425", + "right": "issue:43828", + "accept": false, + "reason": "A torch version compatibility issue is not the same as a model-specific dtype mismatch under autocast." + }, + { + "left": "issue:43257", + "right": "issue:44863", + "accept": false, + "reason": "Both are load failures for large models, but they involve different architectures and different failure mechanisms." + }, + { + "left": "issue:44589", + "right": "issue:44756", + "accept": false, + "reason": "Missing Float8 storage and disabling mmap for OOM are unrelated problems." + }, + { + "left": "issue:44683", + "right": "issue:45823", + "accept": false, + "reason": "Torch flex_attention compilation on >=2.9 and Gemma4 PLE device-map mismatch are different bugs." + }, + { + "left": "issue:39404", + "right": "issue:45701", + "accept": false, + "reason": "Whisper return_language pipeline regression is unrelated to tokenization behavior changes." + }, + { + "left": "issue:43425", + "right": "issue:45823", + "accept": false, + "reason": "General torch incompatibility and Gemma4 device mismatch do not look like the same bug." + }, + { + "left": "issue:43425", + "right": "issue:43716", + "accept": false, + "reason": "Torch incompatibility and Mistral-3 image/text dtype mismatch are different issues." + }, + { + "left": "issue:43611", + "right": "issue:43645", + "accept": false, + "reason": "Both involve custom models, but one is a base_model_prefix loading regression and the other is Jupyter notebook initialization behavior." + }, + { + "left": "issue:43906", + "right": "issue:44530", + "accept": false, + "reason": "An isolated reproduction of another issue and a PagedAttentionCache linear_attention crash are not the same concrete bug." + }, + { + "left": "issue:44530", + "right": "issue:45084", + "accept": false, + "reason": "PagedAttentionCache invalid group type and compiler non-template-node errors are unrelated failures." + }, + { + "left": "issue:39404", + "right": "issue:44671", + "accept": false, + "reason": "Whisper pipeline language return and CamemBERT masked LM prediction drift are unrelated model-specific regressions." + }, + { + "left": "issue:43901", + "right": "issue:44509", + "accept": false, + "reason": "Documentation mismatch about return_all_scores is unrelated to docs mentioning removed pipeline tasks." + }, + { + "left": "issue:44265", + "right": "issue:44898", + "accept": false, + "reason": "torch.export failing on torch_compilable_check is a different issue from Perceiver image classification failing with interpolate_pos_encoding." + }, + { + "left": "issue:44479", + "right": "issue:44805", + "accept": false, + "reason": "A video-input regression for Qwen/VL models is not the same as a generic mask-shape IndexError." + }, + { + "left": "issue:43994", + "right": "issue:44079", + "accept": false, + "reason": "Nonsensical predictions from a specific model are not the same as ModelOutput key assignment logic breaking after None values." + }, + { + "left": "issue:43425", + "right": "issue:44918", + "accept": false, + "reason": "Torch compatibility and Qwen3.5 embedding unpacking are different regressions." + }, + { + "left": "issue:43299", + "right": "issue:43931", + "accept": false, + "reason": "Both are Qwen3-VL loading problems, but the titles indicate different failure modes and not enough evidence of the same root cause." + }, + { + "left": "issue:44589", + "right": "issue:45084", + "accept": false, + "reason": "A storage-object lookup failure and a compiler template-node error are unrelated." + }, + { + "left": "issue:43596", + "right": "issue:44683", + "accept": false, + "reason": "Zero3 init index error with BertModel is unrelated to flex_attention compilation on torch >=2.9." + }, + { + "left": "issue:44164", + "right": "issue:45759", + "accept": false, + "reason": "extra_state save/load handling and unwrapping text_config for composite Qwen configs are different loading/saving bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative. The resize_token_embeddings pair is the only clearly duplicate-looking match, and the other soft-edge verdicts appropriately avoid overclaiming despite superficial subsystem overlap." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 8, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:32090", + "issue:34567", + "issue:36032", + "issue:39404", + "issue:40444", + "issue:40990", + "issue:42915", + "issue:43262", + "issue:43317", + "issue:43377", + "issue:43421", + "issue:43452", + "issue:43482", + "issue:43541", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43643", + "issue:43644", + "issue:43645", + "issue:43646", + "issue:43716", + "issue:43856", + "issue:43883", + "issue:44164", + "issue:44206", + "issue:44220", + "issue:44451", + "issue:44464", + "issue: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:45507", + "issue:45584", + "issue:45859", + "issue:45987" + ], + "soft_pairs": [ + "issue:36032|issue:43452", + "issue:44671|issue:44991", + "issue:32090|issue:39404", + "issue:43856|issue:44756", + "issue:44991|issue:45081", + "issue:44737|issue:45859", + "issue:44164|issue:44479", + "issue:43582|issue:43632", + "issue: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:43643|issue:44530", + "issue:34567|issue:40444", + "issue:43883|issue:44488", + "issue:44811|issue:45584", + "issue:44451|issue:44987", + "issue:43541|issue:45084", + "issue:42915|issue:44530", + "issue:44464|issue:45987", + "issue:43452|issue:43482", + "issue:43644|issue:43646", + "issue:43317|issue:43482", + "issue:44849|issue:45325", + "issue:36032|issue:43645", + "issue:44987|issue:44991", + "issue:45084|issue:45507", + "issue:43883|issue:44991", + "issue:43262|issue:43377" + ], + "prepared_review_unit_hash": "e12f91e6f5805f48aa745d7c27fbb87d4d47f2c0731dfff93e07b6f7c19ea290", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37542, + "estimated_input_tokens": 9386, + "estimated_eval_tokens": 19028 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose set of Transformers v5 regressions and model/tokenizer-loading bugs, but the items are not the same underlying defect. I would keep one broad v5 regression report as the representative issue, and reject all soft duplicate links.", + "confidence": 0.41, + "canonical_issue_reason": "issue:43646 is the broadest and most representative title in the cluster: a generic Transformers v5 custom-model initialization regression. It best matches the cluster\u2019s dominant theme, even though the overall set is still heterogeneous.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43646 is the best single issue to keep because it is broad, central to the v5 regression theme, and not overly tied to one model or one narrow symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36032", + "right": "issue:43452", + "accept": false, + "reason": "Different failure modes: T5 tokenizer attribute conflict vs gguf_file loading path breakage." + }, + { + "left": "issue:44671", + "right": "issue:44991", + "accept": false, + "reason": "Both are model-loading regressions, but one is masked LM prediction quality and the other is tokenizer loading; not the same bug." + }, + { + "left": "issue:32090", + "right": "issue:39404", + "accept": false, + "reason": "Trainer GPU broadcast TypeError and Whisper return_language pipeline regression are unrelated." + }, + { + "left": "issue:43856", + "right": "issue:44756", + "accept": false, + "reason": "Different symptoms and subsystems: Qwen3 MoE training memory usage vs mmap OOM on Strix Halo." + }, + { + "left": "issue:44991", + "right": "issue:45081", + "accept": false, + "reason": "Both affect loading, but one is tokenizer loading for a specific model and the other is a Mistral regex patch crash; distinct code paths." + }, + { + "left": "issue:44737", + "right": "issue:45859", + "accept": false, + "reason": "CPU device-placement bug in XLNet vs Qwen3_5Moe rotary embedding offload compatibility; not the same issue." + }, + { + "left": "issue:44164", + "right": "issue:44479", + "accept": false, + "reason": "save/from_pretrained extra_state handling is unrelated to the Qwen VL video-input regression." + }, + { + "left": "issue:43582", + "right": "issue:43632", + "accept": false, + "reason": "AppleSilicon allocator warmup TypeError and _is_hf_initialized flag regression are separate internal issues." + }, + { + "left": "issue:44661", + "right": "issue:44683", + "accept": false, + "reason": "Tokenizer mapping add-new-model-like failure is unrelated to compiled flex_attention on torch >= 2.9." + }, + { + "left": "issue:44488", + "right": "issue:45081", + "accept": false, + "reason": "Different model families and failure points: generic BERT loading vs Mistral tokenizer regex patch." + }, + { + "left": "issue:44488", + "right": "issue:44987", + "accept": false, + "reason": "Both are load failures, but one is a specific model load issue and the other is a different transformers>=5.1 regression." + }, + { + "left": "issue:43611", + "right": "issue:43644", + "accept": false, + "reason": "Base_model_prefix loading breakage and non-persistent buffer junk are related to v5 internals but are not the same concrete bug." + }, + { + "left": "issue:40990", + "right": "issue:43421", + "accept": false, + "reason": "Perplexity regression on GPT-OSS is not the same as runtime post-processor update behavior in TokenizersBackend." + }, + { + "left": "issue:43716", + "right": "issue:45084", + "accept": false, + "reason": "Image preprocessor/model dtype mismatch is unrelated to the non-template-node compilation error." + }, + { + "left": "issue:44206", + "right": "issue:44220", + "accept": false, + "reason": "These are both audio-feature related, but one is an unsupported center argument regression and the other is a different fbank extraction problem." + }, + { + "left": "issue:36032", + "right": "issue:42915", + "accept": false, + "reason": "Tokenizer load error for T5 is unrelated to Qwen3Moe FineGrainedFP8Config failure." + }, + { + "left": "issue:43643", + "right": "issue:44530", + "accept": false, + "reason": "Missing fields from AutoConfig with trust_remote_code=True is not the same bug as PagedAttentionCache rejecting linear_attention." + }, + { + "left": "issue:34567", + "right": "issue:40444", + "accept": false, + "reason": "TrainerState token counting bug and multi-image IterableDataset finetuning failure are unrelated." + }, + { + "left": "issue:43883", + "right": "issue:44488", + "accept": false, + "reason": "Missing all_tied_weights_keys on Molmo is not the same as loading cjvt/sleng-bert." + }, + { + "left": "issue:44811", + "right": "issue:45584", + "accept": false, + "reason": "Both are Whisper-related, but one ignores skip_special_tokens and the other fails on empty transcription after align_special_tokens; different code paths." + }, + { + "left": "issue:44451", + "right": "issue:44987", + "accept": false, + "reason": "Different affected assets and symptoms: one cannot load a specific model, the other is a broader transformers>=5.1 load regression." + }, + { + "left": "issue:43541", + "right": "issue:45084", + "accept": false, + "reason": "torch dynamo grouped_mm tracing failure is unrelated to template compilation errors." + }, + { + "left": "issue:42915", + "right": "issue:44530", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Qwen3.5 PagedAttentionCache linear_attention crash are different bugs." + }, + { + "left": "issue:44464", + "right": "issue:45987", + "accept": false, + "reason": "Chunked generation inconsistency with compiled forward is not the same as StaticCache.get_seq_length() returning the wrong type." + }, + { + "left": "issue:43452", + "right": "issue:43482", + "accept": false, + "reason": "Both involve loading, but one is gguf_file handling in general and the other is a Qwen2.5-GGUF v5 compatibility regression." + }, + { + "left": "issue:43644", + "right": "issue:43646", + "accept": false, + "reason": "Related v5 breakages, but one is non-persistent buffer initialization and the other is broader custom model initialization; not identical." + }, + { + "left": "issue:43317", + "right": "issue:43482", + "accept": false, + "reason": "gpu+cpu offload dequantized-model loading issue is different from Qwen2.5-GGUF v5 loading failure." + }, + { + "left": "issue:44849", + "right": "issue:45325", + "accept": false, + "reason": "Qwen3.5 hidden-states output bug and Qwen2.5-VL rope-index scaling bug affect different model behaviors." + }, + { + "left": "issue:36032", + "right": "issue:43645", + "accept": false, + "reason": "T5 tokenizer conflict and Jupyter custom-model initialization regression are different problems." + }, + { + "left": "issue:44987", + "right": "issue:44991", + "accept": false, + "reason": "One is a model loading regression, the other is tokenizer loading for a specific checkpoint; same theme but not same bug." + }, + { + "left": "issue:45084", + "right": "issue:45507", + "accept": false, + "reason": "Template-node compilation failure is unrelated to GraniteMoEHybrid calling an invalid method." + }, + { + "left": "issue:43883", + "right": "issue:44991", + "accept": false, + "reason": "Molmo missing attribute and est-roberta tokenizer load failure are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:43377", + "accept": false, + "reason": "Audio chat-template sampling-rate default bug and MIMI padding-mask batch inconsistency are distinct audio 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": 9, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:36032", + "issue:40444", + "issue:40990", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43089", + "issue:43296", + "issue:43322", + "issue:43377", + "issue:43404", + "issue:43425", + "issue:43531", + "issue:43540", + "issue:43611", + "issue:43632", + "issue:43645", + "issue:43646", + "issue:43742", + "issue:43792", + "issue:43819", + "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:44811", + "issue:44877", + "issue:44987", + "issue:45020", + "issue:45081", + "issue:45084", + "issue:45127", + "issue:45216", + "issue:45325", + "issue:45357", + "issue:45542", + "issue:45987" + ], + "soft_pairs": [ + "issue:43089|issue:45987", + "issue:43404|issue:44466", + "issue:43646|issue:44534", + "issue:43645|issue:44534", + "issue:44403|issue:45081", + "issue:43322|issue:43540", + "issue:43994|issue:44610", + "issue:42915|issue:45084", + "issue:43377|issue:43819", + "issue:44220|issue:44683", + "issue:44451|issue:45081", + "issue:44987|issue:45081", + "issue:44265|issue:44610", + "issue:45216|issue:45357", + "issue:42915|issue:44291", + "issue:43632|issue:44756", + "issue:44062|issue:44987", + "issue:36032|issue:43742", + "issue:42915|issue:44589", + "issue:44479|issue:45325", + "issue:43404|issue:45127", + "issue:43296|issue:43531", + "issue:40444|issue:45325", + "issue:36032|issue:41720", + "issue:43531|issue:44877", + "issue:43425|issue:44403", + "issue:43611|issue:44534", + "issue:43632|issue:44534", + "issue:43792|issue:44589", + "issue:42175|issue:45542", + "issue:40990|issue:44811", + "issue:44683|issue:45020" + ], + "prepared_review_unit_hash": "4bc30f0f77c319c7e38aa4f028c5082124840cd3ffe0d37e5bcd0f6cea8a8a19", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36641, + "estimated_input_tokens": 9161, + "estimated_eval_tokens": 18578 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a mixed set of distinct bugs, with one dominant theme around Transformers v5 model-loading/custom-init regressions plus two smaller duplicate-like subclusters (Qwen3.5 checkpoint saving and Qwen VL rope/position-id handling).", + "confidence": 0.67, + "canonical_issue_reason": "Issue 43646 is the broadest and most central representative of the v5 custom-model initialization/loading regression theme; it is more umbrella-like than the narrower notebook/base_model_prefix/_is_hf_initialized variants.", + "canonical_pr_reason": null, + "best_issue_reason": "43646 is the best single issue to anchor the main cluster because it captures the general v5 custom-model initialization breakage without being overly model-specific.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43089", + "right": "issue:45987", + "accept": false, + "reason": "Different problems: generation throughput/sync overhead vs a Cache API return-type contract bug." + }, + { + "left": "issue:43404", + "right": "issue:44466", + "accept": false, + "reason": "Both involve tied weights/lm_head, but one is a Mistral3 model-specific tying bug and the other is a save_pretrained serialization inconsistency; not the same concrete failure." + }, + { + "left": "issue:43646", + "right": "issue:44534", + "accept": false, + "reason": "Custom model initialization failure vs non-persistent buffer corruption; same release family, but different code paths and symptoms." + }, + { + "left": "issue:43645", + "right": "issue:44534", + "accept": false, + "reason": "Jupyter/custom model init regression is unrelated to non-persistent buffer junk filling." + }, + { + "left": "issue:44403", + "right": "issue:45081", + "accept": false, + "reason": "Generic loading noise issue and a Mistral tokenizer patch crash are different bugs." + }, + { + "left": "issue:43322", + "right": "issue:43540", + "accept": false, + "reason": "Different multimodal models and different load/process failures." + }, + { + "left": "issue:43994", + "right": "issue:44610", + "accept": false, + "reason": "Different vision-model issues: nonsensical AutoModel/pipeline outputs vs processor size mismatch." + }, + { + "left": "issue:42915", + "right": "issue:45084", + "accept": false, + "reason": "Qwen3Moe FP8 configuration failure is unrelated to the template-node compilation error." + }, + { + "left": "issue:43377", + "right": "issue:43819", + "accept": false, + "reason": "Both are audio-model correctness bugs, but one is padding-mask batching and the other is STE/latent decoding; not the same bug." + }, + { + "left": "issue:44220", + "right": "issue:44683", + "accept": false, + "reason": "Audio feature extraction bug vs compiled flex_attention incompatibility on newer torch." + }, + { + "left": "issue:44451", + "right": "issue:45081", + "accept": false, + "reason": "Model load failure for ScandiBERT vs tokenizer regex patch crash for Mistral; unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45081", + "accept": false, + "reason": "Different loading regressions affecting different model families and code paths." + }, + { + "left": "issue:44265", + "right": "issue:44610", + "accept": false, + "reason": "torch.export/exportability bug vs OmDet-Turbo processor/image-size mismatch." + }, + { + "left": "issue:45216", + "right": "issue:45357", + "accept": true, + "reason": "Same model family and same `save_pretrained` serialization regression for Qwen3.5; 45357 is a narrower symptom of the checkpoint corruption reported in 45216." + }, + { + "left": "issue:42915", + "right": "issue:44291", + "accept": false, + "reason": "Both are init/loading-related, but Qwen3Moe FP8 config and `_is_hf_initialized` argument handling are distinct bugs." + }, + { + "left": "issue:43632", + "right": "issue:44756", + "accept": false, + "reason": "Different failures: `_is_hf_initialized` initialization behavior vs mmap/OOM policy on Strix Halo." + }, + { + "left": "issue:44062", + "right": "issue:44987", + "accept": false, + "reason": "Tokenizer AddedToken argument duplication vs a model load failure for `physical-intelligence/fast`." + }, + { + "left": "issue:36032", + "right": "issue:43742", + "accept": false, + "reason": "T5 tokenizer special-token conflict and MobileLLM key error are unrelated loading bugs." + }, + { + "left": "issue:42915", + "right": "issue:44589", + "accept": false, + "reason": "Qwen3Moe FP8 config failure is unrelated to the Float8 storage lookup error." + }, + { + "left": "issue:44479", + "right": "issue:45325", + "accept": true, + "reason": "Same Qwen VL rope/temporal position-id regression in v5.3.0; 45325 is the specific still-image manifestation of the broader video-input failure in 44479." + }, + { + "left": "issue:43404", + "right": "issue:45127", + "accept": false, + "reason": "Both touch tied embeddings/weights, but one is Mistral3 lm_head tying and the other is LoRA merge with extended vocabulary; not the same bug." + }, + { + "left": "issue:43296", + "right": "issue:43531", + "accept": false, + "reason": "PaddleOCR-VL load failure and Qwen3-MoE sliding_window behavior are different issues." + }, + { + "left": "issue:40444", + "right": "issue:45325", + "accept": false, + "reason": "IterableDataset multi-image finetuning failure is unrelated to Qwen2.5-VL temporal position-id scaling." + }, + { + "left": "issue:36032", + "right": "issue:41720", + "accept": false, + "reason": "T5 tokenizer conflict and Qwen3 auto device mapping cuda assert are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:44877", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE and strict-config loading for granite_speech are different code paths." + }, + { + "left": "issue:43425", + "right": "issue:44403", + "accept": false, + "reason": "Torch version incompatibility and loading noise are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:44534", + "accept": false, + "reason": "`base_model_prefix` loading breakage and non-persistent buffer junk are separate v5 regressions." + }, + { + "left": "issue:43632", + "right": "issue:44534", + "accept": false, + "reason": "Different regressions: `_is_hf_initialized` flag handling vs non-persistent buffer serialization." + }, + { + "left": "issue:43792", + "right": "issue:44589", + "accept": false, + "reason": "Whisper runtime failure and missing Float8 storage object are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45542", + "accept": false, + "reason": "Both mention TensorFlow/backends, but one is a packaging-extra complaint and the other is a runtime undefined-backend error; not clearly the same bug." + }, + { + "left": "issue:40990", + "right": "issue:44811", + "accept": false, + "reason": "Perplexity regression on GPT-OSS and Whisper batch_decode skip_special_tokens behavior are unrelated." + }, + { + "left": "issue:44683", + "right": "issue:45020", + "accept": false, + "reason": "Compiled flex_attention on torch>=2.9 and remote_code loading regressions are different failures." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "Grounded overall: the main v5 custom-init/loading cluster is supported by the titles, and the two accepted subclusters (45216/45357 and 44479/45325) look like the same bugfix family rather than separate issues. The soft-edge rejections are conservative and don\u2019t overclaim identity from similarity alone." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 10, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:33666", + "issue:35707", + "issue:40444", + "issue:41553", + "issue:41720", + "issue:42175", + "issue:43296", + "issue:43317", + "issue:43377", + "issue:43381", + "issue:43421", + "issue:43482", + "issue:43525", + "issue:43531", + "issue:43582", + "issue:43611", + "issue:43632", + "issue:43644", + "issue:43646", + "issue:43749", + "issue:43792", + "issue:43818", + "issue:43856", + "issue:43957", + "issue:43994", + "issue:44079", + "issue:44265", + "issue:44291", + "issue:44403", + "issue:44462", + "issue:44534", + "issue:44568", + "issue:44589", + "issue:44829", + "issue:44898", + "issue:44991", + "issue:45072", + "issue:45081", + "issue:45325", + "issue:45357", + "issue:45381", + "issue:45479", + "issue:45491", + "issue:45538", + "issue:45663", + "issue:45812", + "issue:45834", + "issue:45923" + ], + "soft_pairs": [ + "issue:44079|issue:45663", + "issue:44403|issue:44991", + "issue:41553|issue:43582", + "issue:43421|issue:44568", + "issue:33666|issue:40444", + "issue:44462|issue:45812", + "issue:43818|issue:45357", + "issue:44829|issue:45479", + "issue:43482|issue:45538", + "issue:43317|issue:43856", + "issue:35707|issue:40444", + "issue:44291|issue:44589", + "issue:43381|issue:45923", + "issue:43482|issue:43525", + "issue:44079|issue:44534", + "issue:43377|issue:45491", + "issue:43531|issue:44291", + "issue:41720|issue:45081", + "issue:43749|issue:44079", + "issue:40444|issue:45381", + "issue:43644|issue:44079", + "issue:44265|issue:45072", + "issue:43632|issue:45072", + "issue:42175|issue:43296", + "issue:45325|issue:45381", + "issue:43611|issue:43646", + "issue:43792|issue:43957", + "issue:43994|issue:44898", + "issue:45325|issue:45834" + ], + "prepared_review_unit_hash": "643939d9cdc5ba5b4775cd1ea5dbae9feec9752c7c57cb4c2f4177bc1feb4ef0", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36602, + "estimated_input_tokens": 9151, + "estimated_eval_tokens": 18558 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: most items are distinct Transformers v5 regressions across loading, tokenization, multimodal position IDs, and training/runtime edge cases. No soft pair looks like a true duplicate, so all soft edges are rejected.", + "confidence": 0.74, + "canonical_issue_reason": "If one issue must anchor the cluster, 43611 is the closest umbrella: it\u2019s a broad Transformers v5 model-loading/init regression and is nearer to the other load/init failures than the specialized tokenizer or multimodal bugs.", + "canonical_pr_reason": null, + "best_issue_reason": "43611 is the best representative of the dominant theme in this set: a generic v5 loading/init regression affecting model creation, which is the broadest and most central issue here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44079", + "right": "issue:45663", + "accept": false, + "reason": "Different bugs: ModelOutput key assignment vs Gemma-4 FSDP2 shared_kv_states rebuild in attention." + }, + { + "left": "issue:44403", + "right": "issue:44991", + "accept": false, + "reason": "Unrelated failures: noisy loading logs vs tokenizer loading regression for EMBEDDIA/est-roberta." + }, + { + "left": "issue:41553", + "right": "issue:43582", + "accept": false, + "reason": "No overlap in code path: bad AutoTokenizer error messaging vs Apple Silicon caching_allocator_warmup TypeError." + }, + { + "left": "issue:43421", + "right": "issue:44568", + "accept": false, + "reason": "Both involve special tokens, but one is runtime post-processor sync and the other is add_special_tokens not inserting BOS/EOS; not the same bug." + }, + { + "left": "issue:33666", + "right": "issue:40444", + "accept": false, + "reason": "Different Qwen2-VL/Qwen2.5-VL training failures with different triggers and inputs; not a duplicate." + }, + { + "left": "issue:44462", + "right": "issue:45812", + "accept": false, + "reason": "Both are AutoTokenizer regressions, but one ignores tokenizer.json while the other produces wrong token IDs for Granite models." + }, + { + "left": "issue:43818", + "right": "issue:45357", + "accept": false, + "reason": "Different model-family bugs: Video-LLaVA tower weights/temporal attention vs Qwen3.5 save_pretrained key names." + }, + { + "left": "issue:44829", + "right": "issue:45479", + "accept": false, + "reason": "Both are degenerate sequence-classification training reports, but one is tied to flash_attention_3 and the other to num_labels=1/problem_type handling." + }, + { + "left": "issue:43482", + "right": "issue:45538", + "accept": false, + "reason": "Different areas entirely: Qwen2.5-GGUF loading vs CLIPTokenizer model_max_length handling." + }, + { + "left": "issue:43317", + "right": "issue:43856", + "accept": false, + "reason": "Load/offload failure on device_map=auto vs Qwen3 MoE training memory inefficiency; not the same concrete bug." + }, + { + "left": "issue:35707", + "right": "issue:40444", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values is unrelated to Qwen2.5-VL multi-image IterableDataset failures." + }, + { + "left": "issue:44291", + "right": "issue:44589", + "accept": false, + "reason": "Both are loading-related v5 regressions, but one is an unexpected _is_hf_initialized argument and the other is a missing Float8 storage object." + }, + { + "left": "issue:43381", + "right": "issue:45923", + "accept": false, + "reason": "Gradient checkpointing eval-mode behavior is distinct from a missing supports_gradient_checkpointing flag on a trust_remote_code variant." + }, + { + "left": "issue:43482", + "right": "issue:43525", + "accept": false, + "reason": "Different model/config issues: Qwen2.5-GGUF load failure vs Llama4Config missing pad_token_id." + }, + { + "left": "issue:44079", + "right": "issue:44534", + "accept": false, + "reason": "Different internal regressions: ModelOutput key assignment vs non-persistent buffer initialization." + }, + { + "left": "issue:43377", + "right": "issue:45491", + "accept": false, + "reason": "Both are batching/masking problems, but they affect different models and different mechanisms (padding mask support vs sliding-window all-padding windows)." + }, + { + "left": "issue:43531", + "right": "issue:44291", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior is unrelated to init_empty_weights/_is_hf_initialized loading errors." + }, + { + "left": "issue:41720", + "right": "issue:45081", + "accept": false, + "reason": "Different subsystems: Qwen3 auto device mapping on A800 vs Mistral tokenizer regex patch crash." + }, + { + "left": "issue:43749", + "right": "issue:44079", + "accept": false, + "reason": "FSDP CPU RAM-efficient loading is a separate loading path from ModelOutput key assignment." + }, + { + "left": "issue:40444", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL multimodal issues, but one is multi-image training input handling and the other is video vision_position_ids." + }, + { + "left": "issue:43644", + "right": "issue:44079", + "accept": false, + "reason": "Distinct v5 regressions: non-persistent buffers being filled with junk is not the same as ModelOutput key assignment." + }, + { + "left": "issue:44265", + "right": "issue:45072", + "accept": false, + "reason": "torch.export exportability issues are unrelated to bfloat16 dtype mismatches in SwitchTransformers/TimmWrapperModel." + }, + { + "left": "issue:43632", + "right": "issue:45072", + "accept": false, + "reason": "_is_hf_initialized breakage is a loading/init regression, while 45072 is a dtype mismatch during inference." + }, + { + "left": "issue:42175", + "right": "issue:43296", + "accept": false, + "reason": "Packaging/dependency omission in pip install is unrelated to the PaddleOCR-VL load failure in vLLM." + }, + { + "left": "issue:45325", + "right": "issue:45381", + "accept": false, + "reason": "Related Qwen2.5-VL position-id bugs, but one is still-image temporal scaling and the other is video vision_position_ids; not the same concrete failure." + }, + { + "left": "issue:43611", + "right": "issue:43646", + "accept": false, + "reason": "Both are v5 init/loading regressions, but one is base_model_prefix loading and the other custom model initialization; too different to merge safely." + }, + { + "left": "issue:43792", + "right": "issue:43957", + "accept": false, + "reason": "Whisper inference failure and torch.device(\"meta\") loading breakage are different bugs." + }, + { + "left": "issue:43994", + "right": "issue:44898", + "accept": false, + "reason": "Different model families and failure modes: SigLIP2 nonsense outputs vs Perceiver non-default resolution failure." + }, + { + "left": "issue:45325", + "right": "issue:45834", + "accept": false, + "reason": "Both touch Qwen2.5-VL, but one is a specific rope-index scaling bug and the other is a long-OCR index error; not enough evidence of the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 11, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:44186", + "issue:44265", + "issue:44291", + "issue:44387", + "issue:44466", + "issue:44479", + "issue:44488", + "issue:44568", + "issue:44610", + "issue:44617", + "issue:44877", + "issue:44898", + "issue:45072", + "issue:45325", + "issue:45584", + "issue:45701", + "issue:45759", + "issue:45800", + "issue:45823", + "issue:45920", + "issue:45923", + "issue:46050" + ], + "soft_pairs": [ + "issue:42915|issue:43482", + "issue:43454|issue:44466", + "issue:43296|issue:45800", + "issue:44387|issue:44617", + "issue:44479|issue:45823", + "issue:43701|issue:45325", + "issue:44291|issue:44877", + "issue:43606|issue:44265", + "issue:43881|issue:44488", + "issue:43262|issue:43746", + "issue:43873|issue:45701", + "issue:44568|issue:45920", + "issue:34634|issue:35707", + "issue:40990|issue:45584", + "issue:44186|issue:44265", + "issue:35707|issue:41720", + "issue:43606|issue:43819", + "issue:41720|issue:45823", + "issue:43454|issue:43550", + "issue:43377|issue:44610", + "issue:43317|issue:43873", + "issue:39692|issue:40444", + "issue:43454|issue:45072", + "issue:43572|issue:45759", + "issue:43377|issue:44898", + "issue:43257|issue:45923", + "issue:43322|issue:46050", + "issue:40990|issue:42947", + "issue:43526|issue:43883" + ], + "prepared_review_unit_hash": "c9aba62c634d781729304c6dec86d10601d79cffb59d20725e399d5a0d8a0a49", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36524, + "estimated_input_tokens": 9131, + "estimated_eval_tokens": 18518 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: most pairs share only broad subsystem overlap (quantization, offloading, multimodal, tokenization, audio) but describe different concrete failures and code paths. No soft pair looks like a true duplicate or mergeable same-fix PR case.", + "confidence": 0.96, + "canonical_issue_reason": "issue:43873 is the broadest representative in the cluster: it\u2019s an open, actively discussed offloading/quantization bug with inbound references and sits near several nearby offload-related reports, even though the cluster overall is not a duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 best captures the most central recurring theme in the cluster (offloading with quantization) and has the strongest evidence of broader impact and follow-up activity.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:42915", + "right": "issue:43482", + "accept": false, + "reason": "Different failures: Qwen3 MoE fine-grained FP8 config vs Qwen2.5-GGUF loading under transformers v5. Same broad model-loading area, but not the same bug or fix." + }, + { + "left": "issue:43454", + "right": "issue:44466", + "accept": false, + "reason": "Both mention tied weights/lm_head, but one is a runtime generation bug in AyaVisionConfig and the other is a serialization inconsistency in v5/main. Different code paths." + }, + { + "left": "issue:43296", + "right": "issue:45800", + "accept": false, + "reason": "PaddleOCR-VL load failure in vLLM/transformers 4.53.0 vs a torch 2.4.1 / transformers 5.8.0 incompatibility. Separate compatibility issues." + }, + { + "left": "issue:44387", + "right": "issue:44617", + "accept": false, + "reason": "Both are OOM-related, but one is int4 CUDA reserved-memory growth in Transformers 5.x and the other is Sam3Video model OOM. Too generic to be the same bug." + }, + { + "left": "issue:44479", + "right": "issue:45823", + "accept": false, + "reason": "Both involve multimodal model/device behavior, but one is a video input regression across Qwen VL variants and the other is a PLE device mismatch during forward on Gemma4. Different failures." + }, + { + "left": "issue:43701", + "right": "issue:45325", + "accept": false, + "reason": "Checkpoint resume key mismatch is unrelated to Qwen2.5-VL rope index scaling. No shared code path." + }, + { + "left": "issue:44291", + "right": "issue:44877", + "accept": false, + "reason": "`init_empty_weights` unexpected argument error vs strict config blocking granite_speech config loading. Both are loading issues, but not the same defect." + }, + { + "left": "issue:43606", + "right": "issue:44265", + "accept": false, + "reason": "CPU offload device mismatch for suno/bark-small vs torch.export failing on `torch_compilable_check`. Different subsystems and fixes." + }, + { + "left": "issue:43881", + "right": "issue:44488", + "accept": false, + "reason": "Generic model loading failure for glm-4v-9b vs inability to load a specific BERT model. No evidence of the same underlying bug." + }, + { + "left": "issue:43262", + "right": "issue:43746", + "accept": false, + "reason": "Audio processor chat-template sample rate default vs GraniteSpeech PEFT local checkpoint loading. Different audio/model-loading issues." + }, + { + "left": "issue:43873", + "right": "issue:45701", + "accept": false, + "reason": "Quantization/offloading behavior vs tokenizer-version-dependent tokenization changes. Overlap is only that both are Transformers regressions." + }, + { + "left": "issue:44568", + "right": "issue:45920", + "accept": false, + "reason": "mDeBERTa special-token handling vs broad AutoTokenizer wrong-token-ID regression across many models. Same area, but 45920 is a wider tokenizer regression, not the same concrete bug." + }, + { + "left": "issue:34634", + "right": "issue:35707", + "accept": false, + "reason": "BarkProcessor voice_preset issue and progressive generation with inputs_embeds/past_key_values are unrelated bugs." + }, + { + "left": "issue:40990", + "right": "issue:45584", + "accept": false, + "reason": "High perplexity on gpt-oss-20b is a model/perplexity issue; Whisper empty transcription after align_special_tokens is a generation edge case. Different code paths." + }, + { + "left": "issue:44186", + "right": "issue:44265", + "accept": false, + "reason": "LayoutLMv2Tokenizer crashes on NER/batching vs torch.export with torch_compilable_check. Both touch tooling, but not the same problem." + }, + { + "left": "issue:35707", + "right": "issue:41720", + "accept": false, + "reason": "Progressive generation with inputs_embeds/past_key_values vs Qwen3 auto device mapping cudaErrorAssert. No meaningful overlap beyond generation/runtime failures." + }, + { + "left": "issue:43606", + "right": "issue:43819", + "accept": false, + "reason": "Bark CPU offload device mismatch vs DAC.from_latents mismatch due to missing STE. Both are model-forward inconsistencies, but different models and root causes." + }, + { + "left": "issue:41720", + "right": "issue:45823", + "accept": false, + "reason": "Qwen3 auto device-map cudaErrorAssert vs Gemma4 PLE device mismatch. Both are device-placement failures, but not the same code path or model family." + }, + { + "left": "issue:43454", + "right": "issue:43550", + "accept": false, + "reason": "AyaVision tied lm_head weights vs Bamba-9B-v2 torch.compile/SDPA failure. Different model internals and failure modes." + }, + { + "left": "issue:43377", + "right": "issue:44610", + "accept": false, + "reason": "MIMI encoder batching/padding-mask bug vs OmDet-Turbo processor/model input-size mismatch. Both are processor/model alignment issues, but not the same defect." + }, + { + "left": "issue:43317", + "right": "issue:43873", + "accept": false, + "reason": "Dequantized model load failure with gpu+cpu offload is adjacent to offloading/quantization, but the reports describe different behaviors and scenarios; not clearly the same underlying bug." + }, + { + "left": "issue:39692", + "right": "issue:40444", + "accept": false, + "reason": "SigLIP2 docs example errors vs Qwen2.5-VL finetuning with iterable multi-image datasets. Both are multimodal, but unrelated concrete bugs." + }, + { + "left": "issue:43454", + "right": "issue:45072", + "accept": false, + "reason": "AyaVision lm_head tying bug vs bfloat16 dtype mismatches in CI for SwitchTransformers/TimmWrapperModel. Different issues." + }, + { + "left": "issue:43572", + "right": "issue:45759", + "accept": false, + "reason": "Missing pad_token_idx in StableLmConfig vs AutoModelForCausalLM.from_config not unwrapping text_config for composite Qwen multimodal configs. Both are config-related, but not the same bug." + }, + { + "left": "issue:43377", + "right": "issue:44898", + "accept": false, + "reason": "MIMI padding-mask batch discrepancy vs Perceiver image classification failing at non-default resolution. Different model families and failures." + }, + { + "left": "issue:43257", + "right": "issue:45923", + "accept": false, + "reason": "Qwen3 MoE loading with accelerate+deepspeed vs missing supports_gradient_checkpointing on a trust_remote_code variant. Both involve Qwen/Nemotron model plumbing, but not the same issue." + }, + { + "left": "issue:43322", + "right": "issue:46050", + "accept": false, + "reason": "Llava Next segmentation fault on load vs sinq quantized model save/load breakage. Different failure class and lifecycle stage." + }, + { + "left": "issue:40990", + "right": "issue:42947", + "accept": false, + "reason": "High perplexity evaluation issue vs gradient checkpointing ineffective with PEFT LoRA. No shared bug mechanism." + }, + { + "left": "issue:43526", + "right": "issue:43883", + "accept": false, + "reason": "BeitImageProcessorFast reduce_labels bug vs Molmo missing all_tied_weights_keys AttributeError. Unrelated processor vs model attribute 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": 12, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "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:45081", + "issue:45161", + "issue:45357", + "issue:45406", + "issue:45632", + "issue:45684", + "issue:45701", + "issue:45704", + "issue:45812", + "issue:45823" + ], + "soft_pairs": [ + "issue:43454|issue:43720", + "issue:44466|issue:45357", + "issue:43296|issue:44877", + "issue:41720|issue:43322", + "issue:43749|issue:44534", + "issue:43881|issue:44991", + "issue:41720|issue:44977", + "issue:42617|issue:42915", + "issue:44008|issue:45823", + "issue:40990|issue:45701", + "issue:43404|issue:45357", + "issue:44186|issue:45072", + "issue:43577|issue:43708", + "issue:44387|issue:45704", + "issue:44877|issue:44991", + "issue:36331|issue:42617", + "issue:43873|issue:44756", + "issue:36331|issue:42915", + "issue:44423|issue:45406", + "issue:45632|issue:45684", + "issue:43450|issue:45701", + "issue:43550|issue:43720", + "issue:41720|issue:42617", + "issue:44479|issue:44918", + "issue:44610|issue:44898", + "issue:44186|issue:44610", + "issue:43550|issue:43606", + "issue:42915|issue:43317", + "issue:43295|issue:45701", + "issue:41720|issue:45161", + "issue:44671|issue:44743", + "issue:43526|issue:43881", + "issue:45081|issue:45812" + ], + "prepared_review_unit_hash": "122c752f4c223b2f07b7de5d0102ff91fe3a89bdee350a7f64293d586cecf754", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37813, + "estimated_input_tokens": 9454, + "estimated_eval_tokens": 19164 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is a loose set of mostly unrelated Transformers issues; despite some superficial keyword overlap around loading, offloading, tokenizers, and multimodal models, none of the soft pairs look like true duplicates. I\u2019d anchor the cluster on the broad offloading/quantization report.", + "confidence": 0.9, + "canonical_issue_reason": "issue:43873 is the broadest and most central report in the set: it is open, has the most discussion/inbound references, and covers a general offloading/quantization failure mode that can absorb nearby loading/offload regressions.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43873 is the best representative issue for the cluster because it is broad, active, and likely to attract related offloading/quantization follow-ups.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43454", + "right": "issue:43720", + "accept": false, + "reason": "Different bugs: tied lm_head weights in AyaVision vs unpacking packed BitNet weights during accelerate loading." + }, + { + "left": "issue:44466", + "right": "issue:45357", + "accept": false, + "reason": "Both mention lm_head/serialization, but one is a device-dependent save_pretrained inconsistency and the other is incorrect visual encoder keys in Qwen3.5." + }, + { + "left": "issue:43296", + "right": "issue:44877", + "accept": false, + "reason": "Different failure surfaces: PaddleOCR-VL load failure in vLLM vs strict config blocking granite_speech config loading." + }, + { + "left": "issue:41720", + "right": "issue:43322", + "accept": false, + "reason": "Different symptoms and models: cuda assert with Qwen3 auto device mapping vs Llava Next segmentation fault." + }, + { + "left": "issue:43749", + "right": "issue:44534", + "accept": false, + "reason": "Both are loading-related, but one is an FSDP CPU RAM efficient loading regression and the other is a v5 non-persistent buffer corruption issue." + }, + { + "left": "issue:43881", + "right": "issue:44991", + "accept": false, + "reason": "Different components: glm-4v-9b model loading vs tokenizer loading for EMBEDDIA/est-roberta." + }, + { + "left": "issue:41720", + "right": "issue:44977", + "accept": false, + "reason": "Same model family, but different concrete bug: device-map CUDA assert vs flash-attention generation regression." + }, + { + "left": "issue:42617", + "right": "issue:42915", + "accept": false, + "reason": "Unrelated issues: a 3d_parallel.py runtime failure vs Qwen3Moe failing with FineGrainedFP8Config." + }, + { + "left": "issue:44008", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma-related, but one is a variable-name collision in forward() and the other is a device-mismatch in PLE forwarding." + }, + { + "left": "issue:40990", + "right": "issue:45701", + "accept": false, + "reason": "High perplexity on gpt-oss-20b is not the same as a transformers-version tokenization change." + }, + { + "left": "issue:43404", + "right": "issue:45357", + "accept": false, + "reason": "Both concern model weights, but one is untied lm_head weights in Mistral3 and the other is incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:44186", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer crashes on NER/padding are unrelated to dtype mismatches in SwitchTransformers/TimmWrapperModel inference." + }, + { + "left": "issue:43577", + "right": "issue:43708", + "accept": false, + "reason": "Different areas: BLIP2 dtype propagation vs trainer resume_from_checkpoint max_steps calculation." + }, + { + "left": "issue:44387", + "right": "issue:45704", + "accept": false, + "reason": "Distinct memory problems: int4 quantization reserved-memory OOM vs apex FusedRMSNorm memory leak." + }, + { + "left": "issue:44877", + "right": "issue:44991", + "accept": false, + "reason": "Different loading paths: strict config rejection vs tokenizer failure for a specific Roberta model." + }, + { + "left": "issue:36331", + "right": "issue:42617", + "accept": false, + "reason": "Training API signature error vs 3d_parallel.py execution failure are unrelated." + }, + { + "left": "issue:43873", + "right": "issue:44756", + "accept": false, + "reason": "Both mention memory/offload, but one is quantization offloading broken and the other is mmap causing OOM on Strix Halo." + }, + { + "left": "issue:36331", + "right": "issue:42915", + "accept": false, + "reason": "Different problems: custom trainer compute_loss signature mismatch vs Qwen3Moe fine-grained FP8 loading failure." + }, + { + "left": "issue:44423", + "right": "issue:45406", + "accept": false, + "reason": "Both affect serve, but one is a continuous-batching multimodal crash and the other is a Gemma4Processor missing _tokenizer attribute." + }, + { + "left": "issue:45632", + "right": "issue:45684", + "accept": false, + "reason": "Unrelated save/cache bugs: trust_remote_code cache-path collision vs read-only permission propagation in save_pretrained." + }, + { + "left": "issue:43450", + "right": "issue:45701", + "accept": false, + "reason": "Video batch-shape regression is unrelated to transformers-version tokenization differences." + }, + { + "left": "issue:43550", + "right": "issue:43720", + "accept": false, + "reason": "Both are model loading/execution issues, but one is torch.compile+SDPA on Bamba and the other is BitNet packed-weight loading." + }, + { + "left": "issue:41720", + "right": "issue:42617", + "accept": false, + "reason": "Different failures: Qwen3 device-map CUDA assert vs inability to run 3d_parallel.py." + }, + { + "left": "issue:44479", + "right": "issue:44918", + "accept": false, + "reason": "Both involve Qwen3.5, but one is a video-input regression and the other is TRL SFT embedding unpacking." + }, + { + "left": "issue:44610", + "right": "issue:44898", + "accept": false, + "reason": "Both are image size/preprocessing complaints, but they involve different models and different expected resolutions/mechanisms." + }, + { + "left": "issue:44186", + "right": "issue:44610", + "accept": false, + "reason": "Tokenizer padding/truncation crashes are not the same bug as OmDet-Turbo producing the wrong image size." + }, + { + "left": "issue:43550", + "right": "issue:43606", + "accept": false, + "reason": "Different device/runtime bugs: torch.compile with SDPA vs CPU-offload device mismatch in bark-small." + }, + { + "left": "issue:42915", + "right": "issue:43317", + "accept": false, + "reason": "Different loading failures: Qwen3Moe FineGrainedFP8Config vs dequantized model load with gpu+cpu offload." + }, + { + "left": "issue:43295", + "right": "issue:45701", + "accept": false, + "reason": "Regression in processor/tokenizer image handling is not the same as a version-dependent tokenization change." + }, + { + "left": "issue:41720", + "right": "issue:45161", + "accept": false, + "reason": "Different Qwen-OSS issues: auto device mapping on A800 vs tensor-parallel not working for GPT-OSS MoE." + }, + { + "left": "issue:44671", + "right": "issue:44743", + "accept": false, + "reason": "Separate model bugs: CamemBERT masked LM predictions vs Qwen3.5 recurrent state reset with cache." + }, + { + "left": "issue:43526", + "right": "issue:43881", + "accept": false, + "reason": "Unrelated processors/models: BeitImageProcessorFast reduce_labels bug vs glm-4v-9b loading failure." + }, + { + "left": "issue:45081", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer-related, but they affect different models and different failure modes; not the same underlying bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 13, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:29159", + "issue:30064", + "issue:33357", + "issue:34567", + "issue:34634", + "issue:35707", + "issue:36010", + "issue:36032", + "issue:41720", + "issue:42490", + "issue:42617", + "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:45406", + "issue:45464", + "issue:45479", + "issue:45701", + "issue:45704", + "issue:45759", + "issue:45812" + ], + "soft_pairs": [ + "issue:43531|issue:43572", + "issue:36010|issue:42617", + "issue:43316|issue:44913", + "issue:43540|issue:43782", + "issue:42915|issue:43322", + "issue:43421|issue:45812", + "issue:44186|issue:44361", + "issue:43388|issue:45059", + "issue:29159|issue:36032", + "issue:45406|issue:45464", + "issue:33357|issue:35707", + "issue:30064|issue:34634", + "issue:41720|issue:43746", + "issue:43421|issue:45701", + "issue:43606|issue:43720", + "issue:43550|issue:43653", + "issue:44492|issue:44509", + "issue:43873|issue:45704", + "issue:43606|issue:43653", + "issue:43531|issue:45759", + "issue:34634|issue:41720", + "issue:42490|issue:43329", + "issue:44514|issue:44545", + "issue:34567|issue:35707", + "issue:44625|issue:45479", + "issue:43421|issue:43927", + "issue:44451|issue:44991", + "issue:44898|issue:45072" + ], + "prepared_review_unit_hash": "18616d84faa2811affa9e5e4159c6d91429f9c401f3286141fb0d8972bef3474", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36173, + "estimated_input_tokens": 9044, + "estimated_eval_tokens": 18344 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly a set of unrelated, model-specific issues. The only clear duplicate pair is issue:44514 and issue:44545, which have nearly identical titles and point to the same Qwen2_5_VLProcessor batched-input crash surface.", + "confidence": 0.86, + "canonical_issue_reason": "issue:44514 is the best canonical issue candidate based on the visible metadata: it has the clearer, more specific title for the shared Qwen2_5_VLProcessor.apply_chat_template batched-input crash when padding=False, and issue:44545 appears to be the same report in near-verbatim form.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44514 is the strongest global issue representative because it most clearly names the concrete failure mode and subsystem, and it is the closest match to its duplicate counterpart.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43531", + "right": "issue:43572", + "accept": false, + "reason": "Both are tokenizer/config bugs, but they involve different models and different fields (sliding_window vs pad_token_idx), so they are not the same defect." + }, + { + "left": "issue:36010", + "right": "issue:42617", + "accept": false, + "reason": "GenerationMixin import failure and a 3d_parallel.py execution problem are different failure modes with no clear shared code path." + }, + { + "left": "issue:43316", + "right": "issue:44913", + "accept": false, + "reason": "Both are config inconsistencies, but they affect different models and different properties, so they do not look like one underlying bug." + }, + { + "left": "issue:43540", + "right": "issue:43782", + "accept": false, + "reason": "These are both Qwen multimodal issues, but one is a video-input ValueError and the other is a from_pretrained weight_only error; different bug surfaces." + }, + { + "left": "issue:42915", + "right": "issue:43322", + "accept": false, + "reason": "Qwen3Moe FP8 configuration failure and Llava Next segmentation fault are unrelated model-loading/runtime issues." + }, + { + "left": "issue:43421", + "right": "issue:45812", + "accept": false, + "reason": "Both involve tokenization, but one is runtime post-processor updates after special-token changes and the other is a tokenizer regression for Granite models; not the same concrete bug." + }, + { + "left": "issue:44186", + "right": "issue:44361", + "accept": false, + "reason": "Tokenizer crashes are too broad here; these affect different tokenizers and task flows without evidence of a shared code path." + }, + { + "left": "issue:43388", + "right": "issue:45059", + "accept": false, + "reason": "Metrics label handling and SAM3 text/bounding-box behavior are unrelated problems." + }, + { + "left": "issue:29159", + "right": "issue:36032", + "accept": false, + "reason": "A slow-vs-fast tokenizer inconsistency is not the same issue as the T5Tokenizer add_special_tokens AttributeError." + }, + { + "left": "issue:45406", + "right": "issue:45464", + "accept": false, + "reason": "These are different API paths: a serve-side processor attribute error versus a streaming chat/completions failure." + }, + { + "left": "issue:33357", + "right": "issue:35707", + "accept": false, + "reason": "A MacOS bus error in CLIP loading and progressive generation with inputs_embeds/past_key_values are unrelated." + }, + { + "left": "issue:30064", + "right": "issue:34634", + "accept": false, + "reason": "Void segmentation-map processing and BarkProcessor voice_preset behavior are different subsystems and bugs." + }, + { + "left": "issue:41720", + "right": "issue:43746", + "accept": false, + "reason": "Auto device mapping cudaErrorAssert and PEFT local-checkpoint loading are different failures." + }, + { + "left": "issue:43421", + "right": "issue:45701", + "accept": false, + "reason": "A runtime tokenizer backend sync bug and a broad version-related tokenization regression are not clearly the same issue." + }, + { + "left": "issue:43606", + "right": "issue:43720", + "accept": false, + "reason": "CPU offload device mismatch and packed-weight unpacking during accelerate loading are distinct bugs." + }, + { + "left": "issue:43550", + "right": "issue:43653", + "accept": false, + "reason": "torch.compile/SDPA failure in Bamba-9B-v2 is unrelated to BigBirdTokenizer special-token registration." + }, + { + "left": "issue:44492", + "right": "issue:44509", + "accept": false, + "reason": "A typo in cache-strategy docs and stale pipeline-task documentation are not the same problem." + }, + { + "left": "issue:43873", + "right": "issue:45704", + "accept": false, + "reason": "Quantization/offloading behavior and a T5 apex FusedRMSNorm memory leak are different issues." + }, + { + "left": "issue:43606", + "right": "issue:43653", + "accept": false, + "reason": "Device mismatch during CPU offload and tokenizer special-token registration/decode output are unrelated." + }, + { + "left": "issue:43531", + "right": "issue:45759", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to text_config unwrapping in composite Qwen multimodal configs." + }, + { + "left": "issue:34634", + "right": "issue:41720", + "accept": false, + "reason": "BarkProcessor voice_preset and Qwen3 auto device mapping errors are different model-specific failures." + }, + { + "left": "issue:42490", + "right": "issue:43329", + "accept": false, + "reason": "A local-path import warning and undefined variables in multimodal token counting are different bugs." + }, + { + "left": "issue:44514", + "right": "issue:44545", + "accept": true, + "reason": "The titles are effectively the same and describe the same Qwen2_5_VLProcessor.apply_chat_template batched-input crash when padding=False." + }, + { + "left": "issue:34567", + "right": "issue:35707", + "accept": false, + "reason": "TrainerState token counting and progressive generation with inputs_embeds/past_key_values are unrelated." + }, + { + "left": "issue:44625", + "right": "issue:45479", + "accept": false, + "reason": "Config propagation in Qwen3.5 and a degenerate zero-loss bug across classification models are not the same concrete defect." + }, + { + "left": "issue:44451", + "right": "issue:44991", + "accept": false, + "reason": "A model-loading failure for one checkpoint and a tokenizer-loading regression for another model are different issues." + }, + { + "left": "issue:44898", + "right": "issue:45072", + "accept": false, + "reason": "Perceiver interpolation failure and dtype mismatches in SwitchTransformers/TimmWrapperModel are unrelated model-specific bugs." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The duplicate call on issue:44514/44545 is well supported by the titles, but the reasoning slightly overreaches the packet by asserting that 44514 is \"reproducible\" and the clear canonical issue without body evidence. With only metadata/titles visible here, that should be stated more conservatively. The soft-edge rejections are mostly cautious, but the overall confidence and canonicality claim are a bit too strong for the evidence shown." + }, + "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": 14, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43531|issue:44589", + "issue:44987|issue:45042", + "issue:43782|issue:45381", + "issue:44077|issue:44625", + "issue:45759|issue:45859", + "issue:44918|issue:44936", + "issue:44062|issue:44451", + "issue:44368|issue:45357", + "issue:45588|issue:45759", + "issue:39692|issue:44610", + "issue:45704|issue:45923", + "issue:43746|issue:46050", + "issue:43421|issue:43653", + "issue:45663|issue:45923", + "issue:42886|issue:44843", + "issue:43381|issue:43844", + "issue:43421|issue:44186", + "issue:43531|issue:43645", + "issue:43994|issue:44671", + "issue:45704|issue:45859", + "issue:41720|issue:45704", + "issue:45081|issue:45356", + "issue:43596|issue:44220", + "issue:43746|issue:43873", + "issue:39692|issue:43873", + "issue:45356|issue:45907", + "issue:40444|issue:43746", + "issue:44220|issue:44661", + "issue:30064|issue:36010", + "issue:43720|issue:43819" + ], + "prepared_review_unit_hash": "1ff08558b0fe80be4d5ee0fda49072cc508380642827cbcebd44802ba170d26c", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 30, + "serialized_chars": 36473, + "estimated_input_tokens": 9119, + "estimated_eval_tokens": 18494 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The candidate pairs are mostly topic-adjacent but not duplicate-level matches. They span unrelated tokenizer, multimodal, quantization, training, and config regressions, with no pair sharing the same concrete bug or code path closely enough to merge.", + "confidence": 0.83, + "canonical_issue_reason": "issue:42886 is the broadest representative of the tokenizer/offline-loading subtheme: it describes a clear user-visible failure, has decent discussion activity, and is more general than the narrower implementation-specific tokenizer reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42886 is the best anchor for this cluster because it is concrete yet broad, making it the most representative issue among otherwise disparate topics.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43531", + "right": "issue:44589", + "accept": false, + "reason": "Different failures in different paths: Qwen3-MoE sliding-window behavior vs a Float8 storage import/type error." + }, + { + "left": "issue:44987", + "right": "issue:45042", + "accept": false, + "reason": "One is a model-loading regression for a specific repo; the other is a PIL image processor dependency regression." + }, + { + "left": "issue:43782", + "right": "issue:45381", + "accept": false, + "reason": "Both involve Qwen VL models, but the bugs differ: weight_only loading vs wrong vision position ids for video input." + }, + { + "left": "issue:44077", + "right": "issue:44625", + "accept": false, + "reason": "Unrelated model/config issues: patchtsmixer post_init handling vs Qwen3.5 num_labels propagation." + }, + { + "left": "issue:45759", + "right": "issue:45859", + "accept": false, + "reason": "Different code paths: composite config unwrapping vs CPU-offload incompatibility in rotary embeddings." + }, + { + "left": "issue:44918", + "right": "issue:44936", + "accept": false, + "reason": "Trainer-related, but not the same bug: embedding unpacking in SFT vs evaluate() failing after train()." + }, + { + "left": "issue:44062", + "right": "issue:44451", + "accept": false, + "reason": "Tokenizer-related, but one is an AddedToken argument error while the other is a model loading failure for ScandiBERT." + }, + { + "left": "issue:44368", + "right": "issue:45357", + "accept": false, + "reason": "A warning about tie_word_embeddings is unrelated to incorrect visual encoder key saving." + }, + { + "left": "issue:45588", + "right": "issue:45759", + "accept": false, + "reason": "Flash-attention sink handling and multimodal config unwrapping are different bugs." + }, + { + "left": "issue:39692", + "right": "issue:44610", + "accept": false, + "reason": "Both mention model/processor mismatches, but they affect different models and different failure modes." + }, + { + "left": "issue:45704", + "right": "issue:45923", + "accept": false, + "reason": "A memory leak in apex RMSNorm is unrelated to a missing gradient-checkpointing flag." + }, + { + "left": "issue:43746", + "right": "issue:46050", + "accept": false, + "reason": "Different load/save regressions in different mechanisms: PEFT local checkpoints vs sinq quantized round-trip." + }, + { + "left": "issue:43421", + "right": "issue:43653", + "accept": false, + "reason": "Both concern special tokens, but one is runtime post-processor updates and the other is BigBird mask-token registration." + }, + { + "left": "issue:45663", + "right": "issue:45923", + "accept": false, + "reason": "Gemma-4 FSDP2 attention state reuse is unrelated to a model capability flag missing on trust_remote_code." + }, + { + "left": "issue:42886", + "right": "issue:44843", + "accept": false, + "reason": "Related offline/tokenizer theme, but not the same concrete bug: generic cache loading failure vs a specific mistral-regex patch call path." + }, + { + "left": "issue:43381", + "right": "issue:43844", + "accept": false, + "reason": "Gradient checkpointing in eval mode is a different issue from abnormal gradient growth under ZeRO-3." + }, + { + "left": "issue:43421", + "right": "issue:44186", + "accept": false, + "reason": "Special-token post-processor behavior does not match a LayoutLMv2Tokenizer crash on NER padding/truncation." + }, + { + "left": "issue:43531", + "right": "issue:43645", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to Jupyter custom-model initialization in Transformers 5.0." + }, + { + "left": "issue:43994", + "right": "issue:44671", + "accept": false, + "reason": "Both are wrong-output regressions, but they affect different models and distinct prediction paths." + }, + { + "left": "issue:45704", + "right": "issue:45859", + "accept": false, + "reason": "A T5 RMSNorm memory leak is unrelated to CPU offload incompatibility in Qwen3_5 MoE rotary embedding." + }, + { + "left": "issue:41720", + "right": "issue:45704", + "accept": false, + "reason": "CUDA assert during Qwen3 auto device mapping is not the same bug as the T5 memory leak." + }, + { + "left": "issue:45081", + "right": "issue:45356", + "accept": false, + "reason": "Both touch fix_mistral_regex, but one crashes with AttributeError while the other reports codec-handling regression and a misleading warning." + }, + { + "left": "issue:43596", + "right": "issue:44220", + "accept": false, + "reason": "BERT zero3 init IndexError and audio fbank feature extraction are unrelated." + }, + { + "left": "issue:43746", + "right": "issue:43873", + "accept": false, + "reason": "Local checkpoint loading with PEFT is not the same as quantization offloading behavior." + }, + { + "left": "issue:39692", + "right": "issue:43873", + "accept": false, + "reason": "A SigLIP2 doc/model mismatch and quantization failure is unrelated to offloading with quantization." + }, + { + "left": "issue:45356", + "right": "issue:45907", + "accept": false, + "reason": "Tokenizer codec/regression issues are unrelated to forwarding the revision argument in list_repo_files." + }, + { + "left": "issue:40444", + "right": "issue:43746", + "accept": false, + "reason": "Multi-image iterable dataset finetuning fails for Qwen2.5-VL; PEFT local checkpoint loading is a different problem." + }, + { + "left": "issue:44220", + "right": "issue:44661", + "accept": false, + "reason": "Audio feature extraction and add-new-model-like mapping logic are different subsystems and bugs." + }, + { + "left": "issue:30064", + "right": "issue:36010", + "accept": false, + "reason": "Image-processor void segmentation maps and a GenerationMixin import error have no common code path." + }, + { + "left": "issue:43720", + "right": "issue:43819", + "accept": false, + "reason": "BitNet packed-weight loading and DAC latent reconstruction/STE behavior are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 15, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43720", + "issue:43756", + "issue:43782", + "issue:43818", + "issue:43819", + "issue:43950", + "issue:44008", + "issue:44062", + "issue:44186", + "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:45685", + "issue:45834", + "issue:45987", + "issue:46032" + ], + "soft_pairs": [ + "issue:43540|issue:44008", + "issue:43653|issue:43720", + "issue:43618|issue:43697", + "issue:45183|issue:45479", + "issue:43388|issue:46032", + "issue:43550|issue:45072", + "issue:43716|issue:44008", + "issue:43295|issue:45834", + "issue:44062|issue:44488", + "issue:43550|issue:43819", + "issue:43335|issue:43756", + "issue:43322|issue:45685", + "issue:43950|issue:45987", + "issue:44186|issue:44898", + "issue:44843|issue:45081", + "issue:43653|issue:43819", + "issue:43531|issue:43643", + "issue:43643|issue:44877", + "issue:43540|issue:45381", + "issue:44448|issue:44671", + "issue:43577|issue:43818", + "issue:30064|issue:36331", + "issue:43421|issue:44811", + "issue:43782|issue:44977", + "issue:44821|issue:45084", + "issue:41720|issue:44991", + "issue:39692|issue:45834" + ], + "prepared_review_unit_hash": "e411b3b124260044dd9bf24b1b5e948f49c46161d9d7b1d1175cf65e03466ab9", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35676, + "estimated_input_tokens": 8919, + "estimated_eval_tokens": 18094 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No duplicate cluster emerges here: the soft pairs span unrelated tokenizer, vision, training, cache, and model-loading regressions. All proposed pairings should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue fits this set; the items are heterogeneous and the soft pairs do not describe the same underlying defect.", + "canonical_pr_reason": "No pull requests are present in the provided cluster.", + "best_issue_reason": "No global best issue: each issue is model- or component-specific, so none cleanly represents the others as a duplicate anchor.", + "best_pr_reason": "No pull requests are present in the provided cluster.", + "soft_edge_verdicts": [ + { + "left": "issue:43540", + "right": "issue:44008", + "accept": false, + "reason": "Different multimodal failure modes: Qwen3OmniMoe video processing vs a Gemma3n forward variable-name collision." + }, + { + "left": "issue:43653", + "right": "issue:43720", + "accept": false, + "reason": "Unrelated code paths: tokenizer special-token registration vs BitNet packed-weight loading in accelerate." + }, + { + "left": "issue:43618", + "right": "issue:43697", + "accept": false, + "reason": "Both are v5 regressions, but one is CLIP output field assignment and the other is RTDetr output divergence." + }, + { + "left": "issue:45183", + "right": "issue:45479", + "accept": false, + "reason": "Different bug classes: validation/error-message logic vs an actual zero-loss classification bug." + }, + { + "left": "issue:43388", + "right": "issue:46032", + "accept": false, + "reason": "Different runtime areas: distributed metric gathering vs Mamba2 cache correctness with seq_len > 1." + }, + { + "left": "issue:43550", + "right": "issue:45072", + "accept": false, + "reason": "Separate failures: torch.compile+SDPA behavior vs dtype mismatch during bfloat16 inference." + }, + { + "left": "issue:43716", + "right": "issue:44008", + "accept": false, + "reason": "Different root causes: image preprocessor/model dtype mismatch vs an AttributeError from a forward variable collision." + }, + { + "left": "issue:43295", + "right": "issue:45834", + "accept": false, + "reason": "Different surfaces: processor/tokenizer regression vs a long-OCR index error in Kosmos2.5." + }, + { + "left": "issue:44062", + "right": "issue:44488", + "accept": false, + "reason": "One is an AddedToken argument collision; the other is a tokenizer load failure for a specific model." + }, + { + "left": "issue:43550", + "right": "issue:43819", + "accept": false, + "reason": "Not the same bug: compile/SDPA failure vs DAC latent decoding not matching forward pass." + }, + { + "left": "issue:43335", + "right": "issue:43756", + "accept": false, + "reason": "Different model-specific configuration mistakes: sparse-layer creation vs RoPE-layer count mismatch." + }, + { + "left": "issue:43322", + "right": "issue:45685", + "accept": false, + "reason": "Completely different domains: Llava Next load crash vs an MPS histogram/int type error in moe code." + }, + { + "left": "issue:43950", + "right": "issue:45987", + "accept": false, + "reason": "Unrelated issues: corrupted non-persistent buffers on load vs return-type contract mismatch in StaticCache." + }, + { + "left": "issue:44186", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer/padding crash vs image-classification interpolation failure; same broad area, different defects." + }, + { + "left": "issue:44843", + "right": "issue:45081", + "accept": false, + "reason": "Same helper name appears, but one is offline model_info access and the other is a backend_tokenizer AttributeError." + }, + { + "left": "issue:43653", + "right": "issue:43819", + "accept": false, + "reason": "Tokenizer special-token handling is unrelated to DAC from_latents / STE matching." + }, + { + "left": "issue:43531", + "right": "issue:43643", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and AutoConfig trust_remote_code field loss are unrelated bugs." + }, + { + "left": "issue:43643", + "right": "issue:44877", + "accept": false, + "reason": "Both concern config loading, but one is missing fields from trust_remote_code and the other is strict config rejection of granite_speech." + }, + { + "left": "issue:43540", + "right": "issue:45381", + "accept": false, + "reason": "Different multimodal issues: video input processing vs wrong vision_position_ids in qwen2.5-vl." + }, + { + "left": "issue:44448", + "right": "issue:44671", + "accept": false, + "reason": "Both are v4/v5 output regressions, but they affect different models and different inference paths." + }, + { + "left": "issue:43577", + "right": "issue:43818", + "accept": false, + "reason": "One is dtype propagation in Blip2; the other is a Video-LLaVA architecture/loading issue." + }, + { + "left": "issue:30064", + "right": "issue:36331", + "accept": false, + "reason": "Image segmentation-map processing and CustomTrainer compute_loss argument handling are unrelated." + }, + { + "left": "issue:43421", + "right": "issue:44811", + "accept": false, + "reason": "Tokenizer post-processor updates and Whisper batch_decode skip_special_tokens are different tokenizer behaviors." + }, + { + "left": "issue:43782", + "right": "issue:44977", + "accept": false, + "reason": "Qwen3VL weight_only loading error vs flash-attention generation bug in Qwen3.5; different failure points." + }, + { + "left": "issue:44821", + "right": "issue:45084", + "accept": false, + "reason": "AutoImageProcessor URL loading and template-node compilation errors are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:44991", + "accept": false, + "reason": "CUDA assert in Qwen3 auto device mapping is not the same defect as tokenizer loading for est-roberta." + }, + { + "left": "issue:39692", + "right": "issue:45834", + "accept": false, + "reason": "SigLIP2 documentation errors and Kosmos2.5 OCR index errors 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": 16, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43697", + "issue:43720", + "issue:43749", + "issue:43782", + "issue:43819", + "issue:43825", + "issue:43827", + "issue:43906", + "issue:43994", + "issue:44060", + "issue:44186", + "issue:44291", + "issue:44368", + "issue:44442", + "issue:44479", + "issue:44485", + "issue:44509", + "issue:44554", + "issue:44561", + "issue:44610", + "issue:44671", + "issue:44805", + "issue:44857", + "issue:44936", + "issue:45020", + "issue:45072", + "issue:45081", + "issue:45084", + "issue:45325", + "issue:45406", + "issue:45538", + "issue:45584", + "issue:45701", + "issue:45834" + ], + "soft_pairs": [ + "issue:44561|issue:45020", + "issue:45584|issue:45834", + "issue:44485|issue:44509", + "issue:43577|issue:43582", + "issue:43644|issue:43749", + "issue:44186|issue:44442", + "issue:44805|issue:44936", + "issue:44479|issue:45834", + "issue:43825|issue:43994", + "issue:43606|issue:45072", + "issue:43825|issue:43827", + "issue:34634|issue:36010", + "issue:44671|issue:45701", + "issue:28218|issue:29159", + "issue:44610|issue:45072", + "issue:43720|issue:45072", + "issue:43782|issue:45325", + "issue:42175|issue:45081", + "issue:43749|issue:44291", + "issue:43582|issue:44857", + "issue:45081|issue:45406", + "issue:43650|issue:43906", + "issue:43262|issue:43819", + "issue:43526|issue:43697", + "issue:44060|issue:44368", + "issue:34634|issue:36331", + "issue:43526|issue:44554", + "issue:45084|issue:45538" + ], + "prepared_review_unit_hash": "4b8ab5bb968b2941bd0b6f7a88c8b2d5f854d9e4338d089bf84613ed476c8d93", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35699, + "estimated_input_tokens": 8925, + "estimated_eval_tokens": 18106 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Almost all pairs describe different models, subsystems, or failure modes. The only likely duplicate is the tokenizer added-token spacing report versus the slow/fast tokenizer inconsistency report.", + "confidence": 0.71, + "canonical_issue_reason": "issue:28218 is the most concrete, reproducible tokenizer bug and best captures the likely duplicate with 29159.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:28218 is the clearest canonical target: specific symptom, easy to reproduce, and more precise than the broader inconsistency report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44561", + "right": "issue:45020", + "accept": false, + "reason": "Related remote_code breakage, but 45020 is a broad umbrella report while 44561 is one specific missing-utility failure; not the same concrete bug." + }, + { + "left": "issue:45584", + "right": "issue:45834", + "accept": false, + "reason": "Whisper empty-transcription failure and Kosmos2.5 long-OCR index error are unrelated models and code paths." + }, + { + "left": "issue:44485", + "right": "issue:44509", + "accept": false, + "reason": "RoPE implementation discussion vs docs still mentioning removed pipeline tasks; different issues." + }, + { + "left": "issue:43577", + "right": "issue:43582", + "accept": false, + "reason": "Blip2 dtype loading bug and Apple Silicon allocator TypeError are unrelated." + }, + { + "left": "issue:43644", + "right": "issue:43749", + "accept": false, + "reason": "Both concern loading, but one is non-persistent buffers in v5 and the other is FSDP CPU RAM efficient loading; different bugs." + }, + { + "left": "issue:44186", + "right": "issue:44442", + "accept": false, + "reason": "Different tokenizer failures on different models; no shared concrete code path." + }, + { + "left": "issue:44805", + "right": "issue:44936", + "accept": false, + "reason": "Mask/index shape error and trainer.evaluate() failing after train are different failure modes." + }, + { + "left": "issue:44479", + "right": "issue:45834", + "accept": false, + "reason": "Video-input regression for Qwen VL models vs Kosmos2.5 OCR long-input error are unrelated." + }, + { + "left": "issue:43825", + "right": "issue:43994", + "accept": false, + "reason": "One is a wrong error message about pipeline support; the other is bad SigLIP2 outputs. Same API surface, different bug." + }, + { + "left": "issue:43606", + "right": "issue:45072", + "accept": false, + "reason": "Bark CPU offload device mismatch and dtype mismatches in different models are unrelated." + }, + { + "left": "issue:43825", + "right": "issue:43827", + "accept": false, + "reason": "Both mention removed pipeline tasks, but one is runtime messaging and the other is documentation cleanup." + }, + { + "left": "issue:34634", + "right": "issue:36010", + "accept": false, + "reason": "BarkProcessor voice_preset bug and GenerationMixin import error are unrelated." + }, + { + "left": "issue:44671", + "right": "issue:45701", + "accept": false, + "reason": "CamemBERT masked-LM regression and broad tokenization-change report are not the same concrete bug." + }, + { + "left": "issue:28218", + "right": "issue:29159", + "accept": true, + "reason": "Both point to tokenizer behavior diverging around added tokens/slow-vs-fast handling; the spacing bug is a concrete manifestation of the inconsistency." + }, + { + "left": "issue:44610", + "right": "issue:45072", + "accept": false, + "reason": "OmDet processor size mismatch and unrelated dtype mismatch in other models are different bugs." + }, + { + "left": "issue:43720", + "right": "issue:45072", + "accept": false, + "reason": "Packed-weight loading failure and dtype mismatch inference failure are unrelated." + }, + { + "left": "issue:43782", + "right": "issue:45325", + "accept": false, + "reason": "Different Qwen VL bugs: one is from_pretrained weight_only loading, the other is rope/index scaling." + }, + { + "left": "issue:42175", + "right": "issue:45081", + "accept": false, + "reason": "Backend extras packaging issue vs Mistral regex patch crash are unrelated." + }, + { + "left": "issue:43749", + "right": "issue:44291", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and init_empty_weights argument handling are different loading regressions." + }, + { + "left": "issue:43582", + "right": "issue:44857", + "accept": false, + "reason": "Apple Silicon allocator TypeError and CUDA AMP loss crash are unrelated." + }, + { + "left": "issue:45081", + "right": "issue:45406", + "accept": false, + "reason": "Tokenizer patch AttributeError and Gemma4Processor missing _tokenizer are different attribute-access bugs." + }, + { + "left": "issue:43650", + "right": "issue:43906", + "accept": false, + "reason": "A placeholder/low-content issue and a reproduction of another issue are not the same bug." + }, + { + "left": "issue:43262", + "right": "issue:43819", + "accept": false, + "reason": "Audio chat-template sample-rate default and DAC from_latents mismatch are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:43697", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug and RTDetrV2 output regression are different issues." + }, + { + "left": "issue:44060", + "right": "issue:44368", + "accept": false, + "reason": "Both involve tied-weights warnings, but one is a specific wrong tie warning and the other is a model-config warning during fine-tuning; not the same bug." + }, + { + "left": "issue:34634", + "right": "issue:36331", + "accept": false, + "reason": "BarkProcessor voice_preset issue and CustomTrainer compute_loss signature mismatch are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:44554", + "accept": false, + "reason": "Image-processor label reduction and MPS attention correctness are unrelated." + }, + { + "left": "issue:45084", + "right": "issue:45538", + "accept": false, + "reason": "Template-compilation TypeError and CLIPTokenizer model_max_length bug are unrelated." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The analyst stays grounded in the packet: the singleton likely duplicate pair (28218/29159) is a defensible tokenizer match, and the rest of the verdicts are conservatively negative given the titles/metadata. No soft-edge acceptance appears overstated from the available evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 17, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43653", + "issue:43696", + "issue:43720", + "issue:43742", + "issue:43746", + "issue:43749", + "issue:43819", + "issue:43854", + "issue:43927", + "issue:43931", + "issue:44479", + "issue:44568", + "issue:44589", + "issue:44610", + "issue:44756", + "issue:44811", + "issue:44912", + "issue:44936", + "issue:44977", + "issue:44991", + "issue:45020", + "issue:45042", + "issue:45356", + "issue:45459", + "issue:45600", + "issue:45759", + "issue:45803", + "issue:45859" + ], + "soft_pairs": [ + "issue:43653|issue:44912", + "issue:43742|issue:45020", + "issue:43329|issue:45600", + "issue:43638|issue:44936", + "issue:45459|issue:45803", + "issue:44589|issue:45042", + "issue:43452|issue:44991", + "issue:43746|issue:44811", + "issue:36010|issue:43645", + "issue:43550|issue:44912", + "issue:38175|issue:44610", + "issue:43819|issue:43927", + "issue:39692|issue:44912", + "issue:43854|issue:44977", + "issue:43257|issue:43329", + "issue:43454|issue:43577", + "issue:43720|issue:44610", + "issue:43531|issue:45042", + "issue:36010|issue:36331", + "issue:43257|issue:43696", + "issue:43262|issue:44811", + "issue:43329|issue:44479", + "issue:42490|issue:43257", + "issue:43749|issue:44756", + "issue:43643|issue:45759", + "issue:44568|issue:45356", + "issue:43329|issue:43450", + "issue:34567|issue:40990", + "issue:43606|issue:45859", + "issue:43257|issue:43931" + ], + "prepared_review_unit_hash": "54a42e6cdd5227154f291a4b2560383dc8ec1f19e35d0a8650b091068c85d406", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36973, + "estimated_input_tokens": 9244, + "estimated_eval_tokens": 18744 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a heterogeneous issue set, but one likely duplicate pair stands out around the Qwen/VL video-input regression; the rest are unrelated bugs across different models and subsystems.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 44479 is the best anchor: it describes a broad, user-facing v5.3.0 video-input regression across several Qwen VL variants and matches the shared multimodal video-path bug.", + "canonical_pr_reason": null, + "best_issue_reason": "44479 is the broadest and most representative issue in the only plausible duplicate thread; it captures the common regression impact more clearly than the narrower root-cause report.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43653", + "right": "issue:44912", + "accept": false, + "reason": "Different problems: BigBirdTokenizer special-token registration vs GPT-oss MXFP4 quantization fallback." + }, + { + "left": "issue:43742", + "right": "issue:45020", + "accept": false, + "reason": "Both involve loading, but one is a specific key error for MobileLLM-125M; the other is a broad remote_code regression." + }, + { + "left": "issue:43329", + "right": "issue:45600", + "accept": false, + "reason": "Different code paths: multimodal video token counting vs stale auto_mappings references to removed configs." + }, + { + "left": "issue:43638", + "right": "issue:44936", + "accept": false, + "reason": "Unrelated training failures: deepspeed zero3/Bert indexing vs trainer.evaluate() after train()." + }, + { + "left": "issue:45459", + "right": "issue:45803", + "accept": false, + "reason": "Both are exception-handling bugs, but in different functions and with different failure modes; not the same change." + }, + { + "left": "issue:44589", + "right": "issue:45042", + "accept": false, + "reason": "Different subsystems: storage lookup for float8 vs image processor torchvision dependency handling." + }, + { + "left": "issue:43452", + "right": "issue:44991", + "accept": false, + "reason": "Both are from_pretrained loading issues, but they affect different mechanisms and models (gguf_file vs tokenizer regression)." + }, + { + "left": "issue:43746", + "right": "issue:44811", + "accept": false, + "reason": "Local checkpoint loading with PEFT adapters is unrelated to Whisper batch_decode skipping special tokens." + }, + { + "left": "issue:36010", + "right": "issue:43645", + "accept": false, + "reason": "ImportError for GenerationMixin is not the same as Jupyter/custom-model initialization breakage in Transformers 5.0." + }, + { + "left": "issue:43550", + "right": "issue:44912", + "accept": false, + "reason": "Different bugs: torch.compile/SDPA failure in Bamba vs quantization fallback in GPT-oss." + }, + { + "left": "issue:38175", + "right": "issue:44610", + "accept": false, + "reason": "Different model/processor issues: zero probabilities in SigLIP2 vs wrong image size in OmDet-Turbo." + }, + { + "left": "issue:43819", + "right": "issue:43927", + "accept": false, + "reason": "DAC forward/from_latents mismatch is unrelated to DiaConfig save/load custom token IDs." + }, + { + "left": "issue:39692", + "right": "issue:44912", + "accept": false, + "reason": "SigLIP2 doc example errors are unrelated to GPT-oss quantization fallback." + }, + { + "left": "issue:43854", + "right": "issue:44977", + "accept": false, + "reason": "Different model-specific generation/loading issues: GLM-4.7-Flash test loading vs Qwen3.5 flash-attention generation." + }, + { + "left": "issue:43257", + "right": "issue:43329", + "accept": false, + "reason": "Qwen3 MoE accelerate/deepspeed weight conversion is unrelated to multimodal video token counting." + }, + { + "left": "issue:43454", + "right": "issue:43577", + "accept": false, + "reason": "AyaVision lm_head tying and Blip2 dtype propagation are distinct model-loading defects." + }, + { + "left": "issue:43720", + "right": "issue:44610", + "accept": false, + "reason": "Packed-weight unpacking during accelerate loading is unrelated to OmDet-Turbo processor image size." + }, + { + "left": "issue:43531", + "right": "issue:45042", + "accept": false, + "reason": "Sliding-window behavior in Qwen3-MoE is unrelated to PIL backend torchvision requirements." + }, + { + "left": "issue:36010", + "right": "issue:36331", + "accept": false, + "reason": "Importing GenerationMixin and CustomTrainer.compute_loss signature mismatch are different bugs." + }, + { + "left": "issue:43257", + "right": "issue:43696", + "accept": false, + "reason": "Qwen3 MoE weight conversion/loading is unrelated to GPT-oss CUDA OOM." + }, + { + "left": "issue:43262", + "right": "issue:44811", + "accept": false, + "reason": "Audio chat-template sample-rate defaults and Whisper batch_decode token skipping are different processor bugs." + }, + { + "left": "issue:43329", + "right": "issue:44479", + "accept": true, + "reason": "Both point to the same multimodal video-input regression; 43329 names the likely root cause in the video branch, while 44479 reports the broad Qwen VL symptom set." + }, + { + "left": "issue:42490", + "right": "issue:43257", + "accept": false, + "reason": "Local-path slash import warning is unrelated to Qwen3 MoE weight conversion/loading." + }, + { + "left": "issue:43749", + "right": "issue:44756", + "accept": false, + "reason": "FSDP CPU RAM efficient loading and Strix Halo mmap OOM are different memory/loading problems." + }, + { + "left": "issue:43643", + "right": "issue:45759", + "accept": false, + "reason": "Both involve config loading, but one is missing fields with trust_remote_code and the other is unwrapping text_config for composite multimodal configs." + }, + { + "left": "issue:44568", + "right": "issue:45356", + "accept": false, + "reason": "Different tokenizer regressions affecting different models and error modes." + }, + { + "left": "issue:43329", + "right": "issue:43450", + "accept": false, + "reason": "The video-token counting bug is not the same as batched video processor shape handling." + }, + { + "left": "issue:34567", + "right": "issue:40990", + "accept": false, + "reason": "Trainer token-count tracking and GPT-oss perplexity are unrelated." + }, + { + "left": "issue:43606", + "right": "issue:45859", + "accept": false, + "reason": "Both mention CPU offload, but they affect different models and code paths." + }, + { + "left": "issue:43257", + "right": "issue:43931", + "accept": false, + "reason": "Weight conversion during accelerate/deepspeed loading is unrelated to Qwen3-VL weight shape mismatch." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative: it identifies one plausible duplicate thread around the Qwen/VL video-input regression while treating the rest as heterogeneous, which matches the titles and soft-similarity evidence. The lone accepted pair (43329 \u2194 44479) is a reasonable same-bugfix match, and the rejected pairs are framed narrowly enough to avoid overclaiming." + }, + "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": 18, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29159", + "issue:33357", + "issue:34567", + "issue:34689", + "issue:38175", + "issue:41720", + "issue:42175", + "issue:42915", + "issue:43257", + "issue:43317", + "issue:43404", + "issue:43452", + "issue:43540", + "issue:43577", + "issue:43582", + "issue:43697", + "issue:43742", + "issue:43756", + "issue:43824", + "issue:43927", + "issue:44062", + "issue:44291", + "issue:44479", + "issue:44488", + "issue:44560", + "issue:44568", + "issue:44737", + "issue:44821", + "issue:44849", + "issue:44898", + "issue:44912", + "issue:44938", + "issue:44987", + "issue:44991", + "issue:45042", + "issue:45072", + "issue:45137", + "issue:45357", + "issue:45381", + "issue:45405", + "issue:45488", + "issue:45538", + "issue:45588", + "issue:45701", + "issue:45907", + "issue:45987" + ], + "soft_pairs": [ + "issue:44568|issue:45488", + "issue:44849|issue:45701", + "issue:29159|issue:34567", + "issue:44291|issue:45042", + "issue:42915|issue:44821", + "issue:43452|issue:43582", + "issue:42915|issue:43742", + "issue:44987|issue:45538", + "issue:43697|issue:45357", + "issue:42915|issue:45538", + "issue:41720|issue:43540", + "issue:33357|issue:34567", + "issue:44479|issue:45381", + "issue:43927|issue:44912", + "issue:42175|issue:43824", + "issue:43317|issue:45538", + "issue:44987|issue:45907", + "issue:43540|issue:44479", + "issue:43540|issue:44560", + "issue:43257|issue:43756", + "issue:43577|issue:45072", + "issue:44062|issue:44991", + "issue:42175|issue:44938", + "issue:42175|issue:44488", + "issue:45137|issue:45405", + "issue:24643|issue:29159", + "issue:43742|issue:44291", + "issue:28282|issue:34689", + "issue:44737|issue:45588", + "issue:43404|issue:45987", + "issue:38175|issue:44898" + ], + "prepared_review_unit_hash": "b2cd293765045c8cd87d83deb9ba62965c9a66699afab61013d219fc61ae83df", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 37041, + "estimated_input_tokens": 9261, + "estimated_eval_tokens": 18778 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly a set of unrelated Transformers issues; only one soft pair looks like a plausible duplicate cluster around Qwen video-input regression. All other pairs are too model-, feature-, or subsystem-specific to merge safely.", + "confidence": 0.71, + "canonical_issue_reason": "Issue 44479 is the best anchor for the only likely duplicate mini-cluster: it describes the broader Qwen video-input regression across multiple Qwen VL/MoE variants, while 45381 is a narrower symptom on qwen2.5-vl. The rest of the set does not form a coherent duplicate cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44479 is the strongest representative issue because it is the broadest and clearest statement of the shared video-input regression, and it subsumes the narrower qwen2.5-vl symptom in 45381.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44568", + "right": "issue:45488", + "accept": false, + "reason": "Both involve tokenizer behavior in v5, but the concrete bugs differ: missing BOS/EOS insertion vs overriding a tokenizer.json pre-tokenizer. Not the same code-path problem." + }, + { + "left": "issue:44849", + "right": "issue:45701", + "accept": false, + "reason": "One is a Qwen3.5 hidden-states bug, the other a broad tokenization-version complaint. Too vague and not the same fix." + }, + { + "left": "issue:29159", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer fast/slow inconsistency and TrainerState token counter updating are unrelated issues." + }, + { + "left": "issue:44291", + "right": "issue:45042", + "accept": false, + "reason": "Different subsystems and failure modes: init_empty_weights argument handling vs PIL image processor dependency requirements." + }, + { + "left": "issue:42915", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3Moe FP8 loading and AutoImageProcessor URL loading are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:43582", + "accept": false, + "reason": "GGUF loading failure and Apple Silicon caching_allocator_warmup TypeError are unrelated code paths." + }, + { + "left": "issue:42915", + "right": "issue:43742", + "accept": false, + "reason": "Different model-loading failures with no shared concrete mechanism." + }, + { + "left": "issue:44987", + "right": "issue:45538", + "accept": false, + "reason": "A model-loading regression and a CLIPTokenizer max-length bug are not the same underlying issue." + }, + { + "left": "issue:43697", + "right": "issue:45357", + "accept": false, + "reason": "Different problems: RTDetrV2 output drift vs incorrect visual encoder keys on save_pretrained." + }, + { + "left": "issue:42915", + "right": "issue:45538", + "accept": false, + "reason": "FP8/Qwen3Moe loading and CLIP tokenizer max-length are unrelated." + }, + { + "left": "issue:41720", + "right": "issue:43540", + "accept": false, + "reason": "Auto device-mapping CUDA assert and Qwen3OmniMoe video processing ValueError are different failures." + }, + { + "left": "issue:33357", + "right": "issue:34567", + "accept": false, + "reason": "Mac bus error on CLIP loading and TrainerState token counting are unrelated." + }, + { + "left": "issue:44479", + "right": "issue:45381", + "accept": true, + "reason": "Both point to the same Qwen video-input regression: 44479 is the broader v5.3.0 regression across Qwen VL/MoE variants, and 45381 is a narrower qwen2.5-vl manifestation involving vision_position_ids." + }, + { + "left": "issue:43927", + "right": "issue:44912", + "accept": false, + "reason": "DiaConfig save/load token-ID loss and MXFP4 quantization fallback are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:43824", + "accept": false, + "reason": "Backend install/dependency packaging issue vs missing Qwen2_5_VL symbol import are different bugs." + }, + { + "left": "issue:43317", + "right": "issue:45538", + "accept": false, + "reason": "device_map/offload dequantization failure and CLIPTokenizer model_max_length are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45907", + "accept": false, + "reason": "Model-loading failure and list_repo_files revision forwarding are unrelated." + }, + { + "left": "issue:43540", + "right": "issue:44479", + "accept": false, + "reason": "Both mention video inputs, but the model families and symptoms differ enough that they do not clearly share the same concrete bug." + }, + { + "left": "issue:43540", + "right": "issue:44560", + "accept": false, + "reason": "Different Qwen video-input failures with different models and error types." + }, + { + "left": "issue:43257", + "right": "issue:43756", + "accept": false, + "reason": "Qwen3 MoE weight conversion with accelerate/deepspeed and Smollm3 RoPE-layer dropping are unrelated." + }, + { + "left": "issue:43577", + "right": "issue:45072", + "accept": false, + "reason": "Both are dtype-related, but they affect different model classes and likely different fixes." + }, + { + "left": "issue:44062", + "right": "issue:44991", + "accept": false, + "reason": "AddedToken duplicate-argument error and tokenizer load failure for EMBEDDIA/est-roberta are not the same bug." + }, + { + "left": "issue:42175", + "right": "issue:44938", + "accept": false, + "reason": "Package/backend install issue and Python 3.14 load failure are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:44488", + "accept": false, + "reason": "TensorFlow backend packaging and loading cjvt/sleng-bert are unrelated." + }, + { + "left": "issue:45137", + "right": "issue:45405", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque error and PEFT minimum-version bump are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44291", + "accept": false, + "reason": "MobileLLM key error and init_empty_weights argument handling are different model-loading problems." + }, + { + "left": "issue:28282", + "right": "issue:34689", + "accept": false, + "reason": "Generic PyTorch ImportError and Llama 3.2 vision model loading breakage are not the same issue." + }, + { + "left": "issue:44737", + "right": "issue:45588", + "accept": false, + "reason": "CPU placement bug in XLNet relative positions and flash_attention s_aux=None crash are unrelated." + }, + { + "left": "issue:43404", + "right": "issue:45987", + "accept": false, + "reason": "Mistral3 weight tying and StaticCache seq-length return type are unrelated code-path issues." + }, + { + "left": "issue:38175", + "right": "issue:44898", + "accept": false, + "reason": "Unexpected zero probabilities in SIGLIP2 and Perceiver non-default-resolution failure are different bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is well grounded in the packet and stays conservative. The only accepted soft pair, issue:44479 vs issue:45381, is plausibly the same Qwen video-input regression with one being a narrower manifestation of the other. The remaining pairs are reasonably rejected as different subsystems, model families, or failure modes, without overstating the evidence." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 19, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:28282", + "issue:29127", + "issue:34567", + "issue:36331", + "issue:41762", + "issue:42915", + "issue:43452", + "issue:43493", + "issue:43526", + "issue:43531", + "issue:43540", + "issue:43582", + "issue:43643", + "issue:43688", + "issue:43697", + "issue:43720", + "issue:43742", + "issue:43749", + "issue:43756", + "issue:43782", + "issue:43906", + "issue:43994", + "issue:44062", + "issue:44291", + "issue:44488", + "issue:44514", + "issue:44589", + "issue:44625", + "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:45600", + "issue:45823", + "issue:45920", + "issue:46121" + ], + "soft_pairs": [ + "issue:43906|issue:45381", + "issue:42915|issue:43749", + "issue:43582|issue:43782", + "issue:28282|issue:29127", + "issue:43526|issue:45479", + "issue:43540|issue:45823", + "issue:44291|issue:44821", + "issue:43756|issue:45325", + "issue:34567|issue:36331", + "issue:43493|issue:43994", + "issue:44913|issue:46121", + "issue:44062|issue:45081", + "issue:44792|issue:44964", + "issue:45357|issue:45920", + "issue:43452|issue:44488", + "issue:45042|issue:45081", + "issue:43742|issue:44589", + "issue:43531|issue:44821", + "issue:43643|issue:45375", + "issue:45405|issue:45600", + "issue:43697|issue:45216", + "issue:41762|issue:44936", + "issue:43688|issue:43749", + "issue:43720|issue:44898", + "issue:44514|issue:45381", + "issue:44625|issue:44877" + ], + "prepared_review_unit_hash": "bf4ffe7573f2f719e4bd1a9e156f88825d7616071d81c09c9d38a2d604ac1383", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 26, + "serialized_chars": 35812, + "estimated_input_tokens": 8953, + "estimated_eval_tokens": 18162 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are not duplicates: they share broad subsystems, but not the same concrete bug. One pair appears to be the same underlying SigLIP2 model-implementation regression.", + "confidence": 0.74, + "canonical_issue_reason": "issue:43493 is the clearest root-cause report: it identifies a concrete SigLIP2 implementation discrepancy against the original JAX model, which can subsume the symptom report in issue:43994.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43493 is the best issue to anchor the cluster because it is specific, technically diagnostic, and points to the underlying defect rather than a downstream symptom.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43906", + "right": "issue:45381", + "accept": false, + "reason": "They are both around Qwen2.5-VL-related behavior, but one is a reproduction of an external issue and the other is a video-position-id bug; no clear same code path." + }, + { + "left": "issue:42915", + "right": "issue:43749", + "accept": false, + "reason": "Different failures in different loading/training paths: FineGrainedFP8Config vs FSDP CPU RAM efficient loading." + }, + { + "left": "issue:43582", + "right": "issue:43782", + "accept": false, + "reason": "AppleSilicon allocator warmup TypeError and Qwen3VL weight_only loading error are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:29127", + "accept": false, + "reason": "One is a missing-PyTorch ImportError, the other is a LayoutLMv3 boxes/clarity issue; unrelated bugs." + }, + { + "left": "issue:43526", + "right": "issue:45479", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and sequence-classification zero-loss behavior are different code paths and symptoms." + }, + { + "left": "issue:43540", + "right": "issue:45823", + "accept": false, + "reason": "Qwen3OmniMoe video-processing error and Gemma4 PLE device mismatch are different model/runtime bugs." + }, + { + "left": "issue:44291", + "right": "issue:44821", + "accept": false, + "reason": "`init_empty_weights` argument handling and loading AutoImageProcessor from URL are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:45325", + "accept": false, + "reason": "Smollm3 RoPE-layer drop and Qwen2.5-VL temporal position-id scaling are different defects." + }, + { + "left": "issue:34567", + "right": "issue:36331", + "accept": false, + "reason": "TrainerState token-count tracking and custom compute_loss kwarg breakage are separate Trainer issues." + }, + { + "left": "issue:43493", + "right": "issue:43994", + "accept": true, + "reason": "Both describe the same SigLIP2 correctness problem: HF outputs diverge from the intended implementation and produce nonsensical results." + }, + { + "left": "issue:44913", + "right": "issue:46121", + "accept": false, + "reason": "One is GPTNeoXConfig persistence of rotary_pct, the other is rope-validation input typing; related area, but not the same bug." + }, + { + "left": "issue:44062", + "right": "issue:45081", + "accept": false, + "reason": "Different tokenizer-loading failures with different error surfaces and code paths." + }, + { + "left": "issue:44792", + "right": "issue:44964", + "accept": false, + "reason": "Janus image-generation test failure and Phi-4 multimodal model loading failure are separate model-specific regressions." + }, + { + "left": "issue:45357", + "right": "issue:45920", + "accept": false, + "reason": "Qwen3.5 save_pretrained visual-key serialization and AutoTokenizer wrong token IDs are different bugs." + }, + { + "left": "issue:43452", + "right": "issue:44488", + "accept": false, + "reason": "Both involve model loading, but one is gguf_file handling for AutoTokenizer/AutoModelForCausalLM and the other is a specific model load failure; not enough evidence of the same root cause." + }, + { + "left": "issue:45042", + "right": "issue:45081", + "accept": false, + "reason": "Image-processor torchvision dependency regression and Mistral regex patch tokenizer crash are unrelated." + }, + { + "left": "issue:43742", + "right": "issue:44589", + "accept": false, + "reason": "MobileLLM key error and missing Float8 storage object are distinct loading/runtime issues." + }, + { + "left": "issue:43531", + "right": "issue:44821", + "accept": false, + "reason": "Qwen3-MoE sliding-window behavior and AutoImageProcessor URL loading are unrelated." + }, + { + "left": "issue:43643", + "right": "issue:45375", + "accept": false, + "reason": "Both are config-field-loss complaints, but they point to different mechanisms and model-specific config schemas; not clearly the same bug." + }, + { + "left": "issue:45405", + "right": "issue:45600", + "accept": false, + "reason": "A version pin to unreleased PEFT and a broken auto_mappings reference are unrelated maintenance issues." + }, + { + "left": "issue:43697", + "right": "issue:45216", + "accept": false, + "reason": "RTDetrV2 output regression and Qwen3.5 save_pretrained regression are different model bugs." + }, + { + "left": "issue:41762", + "right": "issue:44936", + "accept": false, + "reason": "DeepSpeed ZeRO-3 loading crash and trainer.evaluate() after train() failure are separate Trainer/model-loading issues." + }, + { + "left": "issue:43688", + "right": "issue:43749", + "accept": false, + "reason": "Auxiliary-loss normalization and FSDP CPU RAM efficient loading are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:44898", + "accept": false, + "reason": "BitNet packed-weight loading and Perceiver interpolation behavior are different code paths." + }, + { + "left": "issue:44514", + "right": "issue:45381", + "accept": false, + "reason": "Both are Qwen2.5-VL issues, but one is batched chat-template padding and the other is video position-id computation; not the same bug." + }, + { + "left": "issue:44625", + "right": "issue:44877", + "accept": false, + "reason": "Qwen3.5 num_labels propagation and granite_speech config strictness are different config bugs." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster choice is grounded in the packet: issue:43493 is the most root-cause-like SigLIP2 report, and the accepted link to issue:43994 is a plausible same-bug pairing based on the titles. The remaining soft-edge rejections are conservative and do not overclaim duplicate status from broad subsystem similarity alone." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 20, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:24643", + "issue:28282", + "issue:29127", + "issue:29159", + "issue:30064", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:35141", + "issue:38175", + "issue:41720", + "issue:43295", + "issue:43452", + "issue:43577", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43688", + "issue:43697", + "issue:43749", + "issue:43819", + "issue:43906", + "issue:44361", + "issue:44442", + "issue:44464", + "issue:44466", + "issue:44484", + "issue:44493", + "issue:44514", + "issue:44561", + "issue:44568", + "issue:44704", + "issue:44743", + "issue:44811", + "issue:44843", + "issue:44936", + "issue:44991", + "issue:45005", + "issue:45290", + "issue:45356", + "issue:45440", + "issue:45491", + "issue:45701", + "issue:45704", + "issue:45823", + "issue:46002", + "issue:46050" + ], + "soft_pairs": [ + "issue:24643|issue:29127", + "issue:43452|issue:43577", + "issue:43452|issue:44843", + "issue:44514|issue:45290", + "issue:44493|issue:46050", + "issue:43295|issue:45356", + "issue:44484|issue:44704", + "issue:43819|issue:44811", + "issue:44936|issue:45823", + "issue:44466|issue:45005", + "issue:30064|issue:33666", + "issue:44464|issue:44743", + "issue:44361|issue:44442", + "issue:43653|issue:44361", + "issue:41720|issue:43906", + "issue:38175|issue:43295", + "issue:28282|issue:32090", + "issue:28282|issue:33357", + "issue:43653|issue:44568", + "issue:43688|issue:45440", + "issue:43577|issue:43697", + "issue:45491|issue:46002", + "issue:43688|issue:45704", + "issue:28282|issue:35141", + "issue:44843|issue:44991", + "issue:43688|issue:45701", + "issue:43643|issue:44561", + "issue:43646|issue:43749", + "issue:29159|issue:33357" + ], + "prepared_review_unit_hash": "80a4485525d787afd03c01c2afe45959abb792a70f4609ee908ef6315e3a3c28", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 29, + "serialized_chars": 35956, + "estimated_input_tokens": 8989, + "estimated_eval_tokens": 18234 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is largely a grab-bag of unrelated bug reports spanning tokenizers, processors, training, serialization, and model-specific regressions. I don\u2019t see any pair that clearly meets the bar for being the same underlying bug/change.", + "confidence": 0.88, + "canonical_issue_reason": "No clear canonical issue: the items are heterogeneous, and no single report cleanly represents a true duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44466 is the most concrete and actionable report in the set, with a specific v5 tied-weight serialization regression; if one issue must represent the cluster, this is the closest fit.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:24643", + "right": "issue:29127", + "accept": false, + "reason": "Different failures in different subsystems: DeepSpeed training weight-shape error vs LayoutLMv3 box-content error." + }, + { + "left": "issue:43452", + "right": "issue:43577", + "accept": false, + "reason": "Both involve model loading, but one is a gguf/AutoTokenizer breakage and the other is an fp16/bfloat16 dtype bug in BLIP-2; not the same code-path." + }, + { + "left": "issue:43452", + "right": "issue:44843", + "accept": false, + "reason": "Both touch from_pretrained, but one is gguf_file handling and the other is offline Mistral regex/model_info behavior; distinct bugs." + }, + { + "left": "issue:44514", + "right": "issue:45290", + "accept": false, + "reason": "Both are chat-template related, but one is batched padding behavior and the other is tool-call assistant-message handling; too different to merge." + }, + { + "left": "issue:44493", + "right": "issue:46050", + "accept": false, + "reason": "Unexpected position-id keys vs quantized save/load corruption are unrelated symptoms and code paths." + }, + { + "left": "issue:43295", + "right": "issue:45356", + "accept": false, + "reason": "Both are regression-style processor/tokenizer reports, but they describe different APIs and failures, not one concrete bug." + }, + { + "left": "issue:44484", + "right": "issue:44704", + "accept": false, + "reason": "Shard-size default behavior and AutoProcessor kwarg forwarding are unrelated issues." + }, + { + "left": "issue:43819", + "right": "issue:44811", + "accept": false, + "reason": "DAC from_latents parity bug vs Whisper batch_decode skip_special_tokens handling are different model components and problems." + }, + { + "left": "issue:44936", + "right": "issue:45823", + "accept": false, + "reason": "Trainer evaluate-after-train failure and Gemma4 device-map mismatch are separate runtime errors with different causes." + }, + { + "left": "issue:44466", + "right": "issue:45005", + "accept": false, + "reason": "Both mention tied weights, but one is a device-dependent serialization regression while the other is a broader translation-model tied-weights complaint; not clearly the same bug." + }, + { + "left": "issue:30064", + "right": "issue:33666", + "accept": false, + "reason": "Void segmentation map processing and Qwen2-VL multi-GPU training are unrelated image/model-training issues." + }, + { + "left": "issue:44464", + "right": "issue:44743", + "accept": false, + "reason": "Chunked generation with compiled forward and Qwen3 recurrent-state reset are different generation/runtime bugs." + }, + { + "left": "issue:44361", + "right": "issue:44442", + "accept": false, + "reason": "MLukeTokenizer AttributeError and FastSpeech2ConformerTokenizer loading failure are different tokenizer implementations and failures." + }, + { + "left": "issue:43653", + "right": "issue:44361", + "accept": false, + "reason": "BigBird special-token decode issue vs MLukeTokenizer task AttributeError are unrelated tokenizer bugs." + }, + { + "left": "issue:41720", + "right": "issue:43906", + "accept": false, + "reason": "Both reference model execution failures, but one is Qwen3 auto device mapping/cuda assert and the other is just a reproduction of an external issue." + }, + { + "left": "issue:38175", + "right": "issue:43295", + "accept": false, + "reason": "Zero probabilities in SigLIP2 vs processor.tokenizer/images regression are different model and API bugs." + }, + { + "left": "issue:28282", + "right": "issue:32090", + "accept": false, + "reason": "ImportError for missing PyTorch and Trainer _gpu_broadcast_one NoneType error are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:33357", + "accept": false, + "reason": "Missing PyTorch import vs MacOS bus error in CLIP are different failure modes and environments." + }, + { + "left": "issue:43653", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer/special-token related, but BigBird mask-token registration and mdeberta BOS/EOS addition are distinct behaviors." + }, + { + "left": "issue:43688", + "right": "issue:45440", + "accept": false, + "reason": "Aux-loss normalization in MoE models and DeepSeekV3 implementation divergence are not the same code-path problem." + }, + { + "left": "issue:43577", + "right": "issue:43697", + "accept": false, + "reason": "BLIP-2 dtype propagation and RTDetrV2 output drift under v5 are unrelated model-specific regressions." + }, + { + "left": "issue:45491", + "right": "issue:46002", + "accept": false, + "reason": "Gemma3 NaN embeddings with mixed-length batching and classification-head NaNs after checkout/init are different initialization/attention issues." + }, + { + "left": "issue:43688", + "right": "issue:45704", + "accept": false, + "reason": "MoE auxiliary loss normalization and T5/apex FusedRMSNorm memory leak are unrelated bugs." + }, + { + "left": "issue:28282", + "right": "issue:35141", + "accept": false, + "reason": "PyTorch import failure and token-embedding resize/post_init reinitialization are different layers of the stack." + }, + { + "left": "issue:44843", + "right": "issue:44991", + "accept": false, + "reason": "Offline model_info call in AutoTokenizer and tokenizer loading failure for a specific model are not the same underlying defect." + }, + { + "left": "issue:43688", + "right": "issue:45701", + "accept": false, + "reason": "MoE auxiliary-loss normalization and a broad tokenization-change report are too different to be duplicates." + }, + { + "left": "issue:43643", + "right": "issue:44561", + "accept": false, + "reason": "Missing fields from AutoConfig with trust_remote_code and removal of is_torch_fx_available breaking remote-code models are separate compatibility issues." + }, + { + "left": "issue:43646", + "right": "issue:43749", + "accept": false, + "reason": "Custom model initialization breakage in v5 and FSDP_CPU_RAM_EFFICIENT_LOADING failure are different migration/runtime problems." + }, + { + "left": "issue:29159", + "right": "issue:33357", + "accept": false, + "reason": "Slow-vs-fast tokenizer inconsistency and MacOS CLIP bus error are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 21, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:28218", + "issue:32090", + "issue:33357", + "issue:33666", + "issue:34567", + "issue:36010", + "issue:36331", + "issue:41720", + "issue:43122", + "issue:43262", + "issue:43278", + "issue:43295", + "issue:43452", + "issue:43493", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43550", + "issue:43572", + "issue:43577", + "issue:43606", + "issue:43653", + "issue:43688", + "issue:43720", + "issue:43742", + "issue:43756", + "issue:43825", + "issue:43906", + "issue:43927", + "issue:44361", + "issue:44451", + "issue:44464", + "issue:44466", + "issue:44479", + "issue:44554", + "issue:44568", + "issue:44610", + "issue:44625", + "issue:44821", + "issue:44849", + "issue:44913", + "issue:44977", + "issue:45357", + "issue:45588", + "issue:45812", + "issue:45823", + "issue:46002", + "issue:46050" + ], + "soft_pairs": [ + "issue:43531|issue:43742", + "issue:43653|issue:43927", + "issue:44464|issue:44977", + "issue:43577|issue:45588", + "issue:43550|issue:43927", + "issue:44479|issue:44625", + "issue:43550|issue:44361", + "issue:43577|issue:44554", + "issue:43122|issue:43906", + "issue:41720|issue:44464", + "issue:32090|issue:34567", + "issue:43504|issue:43720", + "issue:43504|issue:43550", + "issue:43504|issue:44610", + "issue:44466|issue:46050", + "issue:43742|issue:44821", + "issue:32090|issue:33666", + "issue:33666|issue:36010", + "issue:43452|issue:44451", + "issue:45823|issue:46050", + "issue:43278|issue:43688", + "issue:44849|issue:45357", + "issue:43525|issue:43572", + "issue:43606|issue:43927", + "issue:43720|issue:43927", + "issue:43262|issue:43295", + "issue:44625|issue:46002", + "issue:43493|issue:43825", + "issue:44568|issue:45812", + "issue:43756|issue:44913", + "issue:33666|issue:36331", + "issue:43720|issue:44361", + "issue:28218|issue:33357" + ], + "prepared_review_unit_hash": "81ee92b2c9418d7f42d5d8f83cbc0758dda08a5e2aecac9e60b0cf617d593774", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 33, + "serialized_chars": 37639, + "estimated_input_tokens": 9410, + "estimated_eval_tokens": 19076 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous, but the strongest common thread is v5-era tokenizer/config regressions; most proposed soft matches are actually different model-specific loading/training/runtime bugs.", + "confidence": 0.84, + "canonical_issue_reason": "issue:43122 is the broadest tokenizer regression here: it describes a version-to-version tokenization change with the same tokenizer, which makes it the best umbrella issue for narrower tokenizer symptoms.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45812 is the strongest standalone issue in the set: it names a concrete, high-impact AutoTokenizer regression across a whole model family and already has inbound references.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43531", + "right": "issue:43742", + "accept": false, + "reason": "Different bugs: Qwen3-MoE sliding-window behavior vs MobileLLM-125M load key error." + }, + { + "left": "issue:43653", + "right": "issue:43927", + "accept": false, + "reason": "Both involve token-related problems, but one is BigBird special-token registration and the other is DiaConfig save/load custom token IDs; not the same bug." + }, + { + "left": "issue:44464", + "right": "issue:44977", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency is separate from a flash-attention generation failure in Qwen3.5." + }, + { + "left": "issue:43577", + "right": "issue:45588", + "accept": false, + "reason": "Unrelated issues: dtype propagation in BLIP2 loading vs flash_attention.py crashing on s_aux=None." + }, + { + "left": "issue:43550", + "right": "issue:43927", + "accept": false, + "reason": "Bamba torch.compile/SDPA failure is unrelated to DiaConfig token ID loss on save/load." + }, + { + "left": "issue:44479", + "right": "issue:44625", + "accept": false, + "reason": "Video-input regression for Qwen VL models is not the same as num_labels propagation between configs." + }, + { + "left": "issue:43550", + "right": "issue:44361", + "accept": false, + "reason": "Different failure modes and models: Bamba compile/SDPA vs MLukeTokenizer AttributeError on tasks." + }, + { + "left": "issue:43577", + "right": "issue:44554", + "accept": false, + "reason": "BLIP2 dtype handling and MPS attention correctness are unrelated code paths." + }, + { + "left": "issue:43122", + "right": "issue:43906", + "accept": false, + "reason": "43906 is only an isolated reproduction reference; insufficient evidence that it is the same underlying tokenizer regression." + }, + { + "left": "issue:41720", + "right": "issue:44464", + "accept": false, + "reason": "Qwen3 auto device-map cuda assert and compiled-forward generation inconsistency are different bugs." + }, + { + "left": "issue:32090", + "right": "issue:34567", + "accept": false, + "reason": "Trainer GPU broadcast NoneType error and TrainerState token counter not updating are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:43720", + "accept": false, + "reason": "Legacy field loading in BeitForSemanticSegmentation is not the same as BitNet packed-weight unpacking during accelerate loading." + }, + { + "left": "issue:43504", + "right": "issue:43550", + "accept": false, + "reason": "Different model-loading paths and failures; no shared underlying defect is evident." + }, + { + "left": "issue:43504", + "right": "issue:44610", + "accept": false, + "reason": "Pretrained loading legacy-field issue is unrelated to OmDet-Turbo processor image-size mismatch." + }, + { + "left": "issue:44466", + "right": "issue:46050", + "accept": false, + "reason": "Both touch serialization, but one is lm_head.weight device-dependent save behavior and the other is sinq quantization save/load breakage." + }, + { + "left": "issue:43742", + "right": "issue:44821", + "accept": false, + "reason": "MobileLLM key error and AutoImageProcessor-from-URL loading are different loader failures." + }, + { + "left": "issue:32090", + "right": "issue:33666", + "accept": false, + "reason": "Trainer broadcast error is unrelated to Qwen2-VL multi-GPU training support." + }, + { + "left": "issue:33666", + "right": "issue:36010", + "accept": false, + "reason": "Multi-GPU training for Qwen2-VL and GenerationMixin import failure are unrelated." + }, + { + "left": "issue:43452", + "right": "issue:44451", + "accept": false, + "reason": "gguf_file loading breakage is not the same as a specific model that cannot be loaded." + }, + { + "left": "issue:45823", + "right": "issue:46050", + "accept": false, + "reason": "Gemma4 device-mapping mismatch is unrelated to sinq quantized save/load corruption." + }, + { + "left": "issue:43278", + "right": "issue:43688", + "accept": false, + "reason": "Embedding dtype drift during evaluate is unrelated to auxiliary-loss normalization in MoE models." + }, + { + "left": "issue:44849", + "right": "issue:45357", + "accept": false, + "reason": "Qwen3.5 hidden-states behavior and incorrect visual-encoder save keys are different regressions." + }, + { + "left": "issue:43525", + "right": "issue:43572", + "accept": false, + "reason": "Both mention missing pad-token fields, but they are different configs and different attribute names; not enough to treat as one bug." + }, + { + "left": "issue:43606", + "right": "issue:43927", + "accept": false, + "reason": "CPU offload device mismatch and DiaConfig token-ID loss are unrelated." + }, + { + "left": "issue:43720", + "right": "issue:43927", + "accept": false, + "reason": "Packed-weight unpacking during accelerate loading is unrelated to config token-ID persistence." + }, + { + "left": "issue:43262", + "right": "issue:43295", + "accept": false, + "reason": "Audio chat-template sampling-rate default and processor/tokenizer API regression are separate processor issues." + }, + { + "left": "issue:44625", + "right": "issue:46002", + "accept": false, + "reason": "num_labels propagation and NaNs in classification heads are different problems." + }, + { + "left": "issue:43493", + "right": "issue:43825", + "accept": false, + "reason": "SigLIP2 implementation discrepancy is unrelated to pipeline translation-task messaging." + }, + { + "left": "issue:44568", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenization regressions, but one is mdeberta special-token insertion and the other is Granite token ID mapping; different concrete bugs." + }, + { + "left": "issue:43720", + "right": "issue:44361", + "accept": false, + "reason": "Different loading paths and failures: BitNet accelerate unpacking vs MLukeTokenizer AttributeError." + }, + { + "left": "issue:28218", + "right": "issue:33357", + "accept": false, + "reason": "Tokenizer added-space behavior and MacOS CLIP bus error are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 22, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:44589", + "issue:44734", + "issue:44792", + "issue:44843", + "issue:44964", + "issue:44977", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45405", + "issue:45406", + "issue:45520", + "issue:45584", + "issue:45759", + "issue:45803", + "issue:45823", + "issue:45987", + "issue:46121" + ], + "soft_pairs": [ + "issue:44964|issue:45405", + "issue:39692|issue:43493", + "issue:43540|issue:44977", + "issue:43493|issue:43844", + "issue:43278|issue:43844", + "issue:43334|issue:43572", + "issue:45803|issue:45987", + "issue:45200|issue:45229", + "issue:44488|issue:44843", + "issue:43756|issue:45083", + "issue:43257|issue:43299", + "issue:43334|issue:45759", + "issue:39692|issue:43295", + "issue:43824|issue:44589", + "issue:45229|issue:45823", + "issue:43572|issue:44291", + "issue:43646|issue:44466", + "issue:44589|issue:45520", + "issue:33357|issue:33666", + "issue:43504|issue:43606", + "issue:43927|issue:44361", + "issue:44464|issue:45584", + "issue:43504|issue:43653", + "issue:43606|issue:44112", + "issue:43575|issue:45405", + "issue:43994|issue:44493", + "issue:44792|issue:45405", + "issue:43278|issue:43701", + "issue:44423|issue:44734", + "issue:45406|issue:46121" + ], + "prepared_review_unit_hash": "4c7e0bfe170f5a055da0aaa5fd766cf4bde154a0ddf1f1b42f7753c796411ba5", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36947, + "estimated_input_tokens": 9237, + "estimated_eval_tokens": 18730 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: most items are unrelated model-loading, config, serving, or training bugs, and the soft pairs are topical similarities rather than true duplicates. No single issue or PR cleanly represents the cluster.", + "confidence": 0.99, + "canonical_issue_reason": "No canonical issue: the items span many distinct models and failure modes, so there is no shared underlying bug to collapse onto one issue.", + "canonical_pr_reason": "No PRs are present in the cluster.", + "best_issue_reason": "No global best issue: any one item would be too narrow and would misrepresent the cluster\u2019s mixed bug themes.", + "best_pr_reason": "No PRs are present in the cluster.", + "soft_edge_verdicts": [ + { + "left": "issue:44964", + "right": "issue:45405", + "accept": false, + "reason": "Both are loading-related, but one is a Phi-4 multimodal model load failure and the other is a PEFT version pin/release mismatch; different root causes." + }, + { + "left": "issue:39692", + "right": "issue:43493", + "accept": false, + "reason": "Same model family, but one is a documentation/example problem while the other is an HF-vs-JAX implementation discrepancy; not the same bug." + }, + { + "left": "issue:43540", + "right": "issue:44977", + "accept": false, + "reason": "Both involve Qwen multimodal models, but the failures are different: video input validation in Qwen3OmniMoe vs flash-attention generation behavior in Qwen3.5." + }, + { + "left": "issue:43493", + "right": "issue:43844", + "accept": false, + "reason": "SigLIP2 discrepancy vs DeepSpeed gradient explosion are different code paths and symptoms; no shared underlying defect." + }, + { + "left": "issue:43278", + "right": "issue:43844", + "accept": false, + "reason": "Embedding dtype regression during eval is unrelated to the ZeRO-3 gradient anomaly." + }, + { + "left": "issue:43334", + "right": "issue:43572", + "accept": false, + "reason": "Both are config attribute issues, but they affect different models and missing fields; not the same problem." + }, + { + "left": "issue:45803", + "right": "issue:45987", + "accept": false, + "reason": "Both touch utility contracts, but one is a bare-except bug in Fuyu tensor conversion and the other is a StaticCache return-type mismatch." + }, + { + "left": "issue:45200", + "right": "issue:45229", + "accept": false, + "reason": "Gemma4 mm_token_type_ids defaulting and Gemma4 multi-GPU OOM are separate issues with different failure modes." + }, + { + "left": "issue:44488", + "right": "issue:44843", + "accept": false, + "reason": "A model load failure and an offline-mode tokenizer hub call are only loosely related; different underlying causes." + }, + { + "left": "issue:43756", + "right": "issue:45083", + "accept": false, + "reason": "Both mention RoPE-related internals, but one is about missing RoPE layers in Smollm3 and the other is a helper length computation in qwen3_omni_moe." + }, + { + "left": "issue:43257", + "right": "issue:43299", + "accept": false, + "reason": "Both concern Qwen3 MOE loading, but one is accelerate+deepspeed weight conversion and the other is a dev-version load regression; not the same concrete bug." + }, + { + "left": "issue:43334", + "right": "issue:45759", + "accept": false, + "reason": "Both are multimodal config issues, but one is missing pad_token_id on Qwen3-VL text config and the other is failure to unwrap text_config in AutoModelForCausalLM.from_config." + }, + { + "left": "issue:39692", + "right": "issue:43295", + "accept": false, + "reason": "Documentation/example errors for SigLIP2 are unrelated to the processor.tokenizer regression in a separate multimodal code path." + }, + { + "left": "issue:43824", + "right": "issue:44589", + "accept": false, + "reason": "ImportError for Qwen2_5_VL and missing Float8 storage support are unrelated issues." + }, + { + "left": "issue:45229", + "right": "issue:45823", + "accept": false, + "reason": "Both are Gemma4-related, but one is an inference OOM and the other is a device mismatch with device_map='auto'; different bugs." + }, + { + "left": "issue:43572", + "right": "issue:44291", + "accept": false, + "reason": "StableLmConfig missing pad_token_idx and init_empty_weights receiving _is_hf_initialized are distinct initialization/config regressions." + }, + { + "left": "issue:43646", + "right": "issue:44466", + "accept": false, + "reason": "Custom model initialization breakage and lm_head serialization inconsistency are different bugs in different parts of the stack." + }, + { + "left": "issue:44589", + "right": "issue:45520", + "accept": false, + "reason": "Float8 storage lookup failure and flash_attn import_utils KeyError are unrelated environment/import problems." + }, + { + "left": "issue:33357", + "right": "issue:33666", + "accept": false, + "reason": "MacOS bus error with CLIP and Qwen2-VL multi-GPU training are completely different issues." + }, + { + "left": "issue:43504", + "right": "issue:43606", + "accept": false, + "reason": "BEiT legacy field loading and bark-small CPU offload device mismatch are unrelated model-specific bugs." + }, + { + "left": "issue:43927", + "right": "issue:44361", + "accept": false, + "reason": "DiaConfig token ID persistence and MLukeTokenizer AttributeError affect different components and error modes." + }, + { + "left": "issue:44464", + "right": "issue:45584", + "accept": false, + "reason": "Chunked generation inconsistency with compiled forward is not the same as Whisper empty-transcription failure after align_special_tokens." + }, + { + "left": "issue:43504", + "right": "issue:43653", + "accept": false, + "reason": "Both are loading/tokenizer-related, but BEiT legacy-field loading and BigBirdTokenizer special-token registration are separate bugs." + }, + { + "left": "issue:43606", + "right": "issue:44112", + "accept": false, + "reason": "A real CPU-offload device mismatch is not the same as a stale CI test failure." + }, + { + "left": "issue:43575", + "right": "issue:45405", + "accept": false, + "reason": "Qwen2-57B TP OOM and a PEFT version bump are unrelated." + }, + { + "left": "issue:43994", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 producing nonsensical outputs and unexpected position-id keys across many models are different regression classes." + }, + { + "left": "issue:44792", + "right": "issue:45405", + "accept": false, + "reason": "A janus test failure and a PEFT minimum-version bump are unrelated." + }, + { + "left": "issue:43278", + "right": "issue:43701", + "accept": false, + "reason": "Embedding dtype regression and resume_from_checkpoint key mismatch do not share a concrete code path." + }, + { + "left": "issue:44423", + "right": "issue:44734", + "accept": false, + "reason": "Both are serving-time crashes, but one is a multimodal 'str.to' error and the other is KV-cache tensor indexing; distinct bugs." + }, + { + "left": "issue:45406", + "right": "issue:46121", + "accept": false, + "reason": "Gemma4Processor missing _tokenizer and JSON-loaded rope-key validation TypeError 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": 23, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:15354", + "issue:28218", + "issue:29159", + "issue:29942", + "issue:30333", + "issue:32090", + "issue:34567", + "issue:37428", + "issue:38175", + "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:44315", + "issue:44466", + "issue:44521", + "issue:44554", + "issue:44610", + "issue:44849", + "issue:44898", + "issue:44936", + "issue:44964", + "issue:44977", + "issue:44987", + "issue:45072", + "issue:45084", + "issue:45127", + "issue:45198", + "issue:45200", + "issue:45290", + "issue:45325", + "issue:45356", + "issue:45412", + "issue:45446", + "issue:45520", + "issue:45588", + "issue:45823", + "issue:46050" + ], + "soft_pairs": [ + "issue:43873|issue:46050", + "issue:29942|issue:30333", + "issue:45588|issue:46050", + "issue:42490|issue:43696", + "issue:44521|issue:45290", + "issue:43844|issue:44964", + "issue:28218|issue:34567", + "issue:45084|issue:45520", + "issue:44554|issue:45446", + "issue:29159|issue:32090", + "issue:43540|issue:45200", + "issue:43582|issue:44977", + "issue:15354|issue:38175", + "issue:43232|issue:44849", + "issue:37428|issue:43957", + "issue:45200|issue:45823", + "issue:45072|issue:45198", + "issue:44315|issue:44849", + "issue:43493|issue:43701", + "issue:43295|issue:44610", + "issue:44554|issue:45356", + "issue:43572|issue:44987", + "issue:43572|issue:43957", + "issue:37428|issue:43824", + "issue:43295|issue:43450", + "issue:44898|issue:45412", + "issue:45200|issue:45325", + "issue:44466|issue:45127", + "issue:43844|issue:44936", + "issue:43295|issue:43873" + ], + "prepared_review_unit_hash": "3f97a6ca6c30821392fbd89d94cc395c5194aac6667faf205f373f6deb694b3b", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36785, + "estimated_input_tokens": 9197, + "estimated_eval_tokens": 18650 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is a mixed bag of unrelated Hugging Face/Transformers bug reports; none of the soft pairs are strong enough to merge as duplicates. There are no PRs in the cluster. If a representative issue is needed, issue 44521 is the strongest single candidate because it is specific, reproducible, and has the most substantive discussion among the active reports.", + "confidence": 0.78, + "canonical_issue_reason": "Issue 44521 is the clearest, most actionable open bug report here: it describes a concrete multimodal `apply_chat_template` failure, has strong discussion/inbound references, and is more specific than the broader or more ambiguous issues.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 44521 is the best representative issue overall because it is well-scoped, user-facing, and has enough engagement to serve as a stable canonical reference.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43873", + "right": "issue:46050", + "accept": false, + "reason": "Both mention quantization, but one is an offloading behavior issue and the other is a save/load corruption regression; different code paths and symptoms." + }, + { + "left": "issue:29942", + "right": "issue:30333", + "accept": false, + "reason": "FlashAttention CI failures and MLflow reporting behavior are unrelated subsystems." + }, + { + "left": "issue:45588", + "right": "issue:46050", + "accept": false, + "reason": "A flash-attention crash on `s_aux=None` is unrelated to a quantized model breaking after save/load." + }, + { + "left": "issue:42490", + "right": "issue:43696", + "accept": false, + "reason": "A local-path import warning and a CUDA OOM report are not the same bug." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Both involve `apply_chat_template`, but one is all-zero assistant masks for multimodal inputs while the other crashes on tool-call assistant messages with no content." + }, + { + "left": "issue:43844", + "right": "issue:44964", + "accept": false, + "reason": "ZeRO-3 gradient instability and a Phi-4 multimodal loading failure are unrelated." + }, + { + "left": "issue:28218", + "right": "issue:34567", + "accept": false, + "reason": "Tokenizer spacing after added tokens is unrelated to TrainerState token-count tracking." + }, + { + "left": "issue:45084", + "right": "issue:45520", + "accept": false, + "reason": "A compilation error for non-template nodes is not the same as a Python 3.13 `flash_attn` import key error." + }, + { + "left": "issue:44554", + "right": "issue:45446", + "accept": false, + "reason": "An MPS attention correctness issue and a PyTorch version-check/import guard bug are different problems." + }, + { + "left": "issue:29159", + "right": "issue:32090", + "accept": false, + "reason": "Slow/fast tokenizer inconsistency is unrelated to the `_gpu_broadcast_one` NoneType error in Trainer." + }, + { + "left": "issue:43540", + "right": "issue:45200", + "accept": false, + "reason": "Video-input processing in Qwen3OmniMoe is unrelated to Gemma 4 text-only `mm_token_type_ids` defaults." + }, + { + "left": "issue:43582", + "right": "issue:44977", + "accept": false, + "reason": "Apple Silicon allocator warmup errors and Qwen3.5 flash-attention generation failures are different code paths." + }, + { + "left": "issue:15354", + "right": "issue:38175", + "accept": false, + "reason": "TorchScript GeneratorExp export failure is unrelated to the SigLIP2 zero-probability report." + }, + { + "left": "issue:43232", + "right": "issue:44849", + "accept": false, + "reason": "A generation kwargs sync issue is not the same as the Qwen3.5 `output_hidden_states=True` bug." + }, + { + "left": "issue:37428", + "right": "issue:43957", + "accept": false, + "reason": "A flash-attention import symbol error and a `torch.device('meta')` model-loading regression are distinct failures." + }, + { + "left": "issue:45200", + "right": "issue:45823", + "accept": false, + "reason": "Missing default multimodal token type ids and a Gemma4 PLE device-mapping mismatch are not the same bug." + }, + { + "left": "issue:45072", + "right": "issue:45198", + "accept": false, + "reason": "Dtype mismatches in inference and Wav2Vec2 save/tokenization failures do not share a concrete code-path." + }, + { + "left": "issue:44315", + "right": "issue:44849", + "accept": false, + "reason": "Liger Kernel not being applied with `model_init` is unrelated to the hidden-states bug in Qwen3.5." + }, + { + "left": "issue:43493", + "right": "issue:43701", + "accept": false, + "reason": "A SigLIP2 HF/JAX discrepancy and a resume-checkpoint key mismatch are different issues." + }, + { + "left": "issue:43295", + "right": "issue:44610", + "accept": false, + "reason": "The processor/tokenizer regression and the OmDet-Turbo image-size mismatch are separate preprocessing bugs." + }, + { + "left": "issue:44554", + "right": "issue:45356", + "accept": false, + "reason": "An MPS attention correctness problem is unrelated to the Kimi-K2.5 tokenizer regression." + }, + { + "left": "issue:43572", + "right": "issue:44987", + "accept": false, + "reason": "A missing `pad_token_idx` config field is not the same as a model-loading failure for `physical-intelligence/fast`." + }, + { + "left": "issue:43572", + "right": "issue:43957", + "accept": false, + "reason": "Config defaults and meta-device loading failures are different regression types." + }, + { + "left": "issue:37428", + "right": "issue:43824", + "accept": false, + "reason": "These are both import errors, but they involve different symbols and model families with no shared concrete bug." + }, + { + "left": "issue:43295", + "right": "issue:43450", + "accept": false, + "reason": "A processor/tokenizer regression is not the same as batched video shape handling." + }, + { + "left": "issue:44898", + "right": "issue:45412", + "accept": false, + "reason": "Perceiver interpolation failures and RT-DETR memory-release behavior are unrelated." + }, + { + "left": "issue:45200", + "right": "issue:45325", + "accept": false, + "reason": "Gemma 4 text-only defaults and Qwen2.5-VL rope-index scaling are different multimodal bugs." + }, + { + "left": "issue:44466", + "right": "issue:45127", + "accept": false, + "reason": "Both touch tied weights, but one is serialization/device-dependent state and the other is LoRA merge collapse with extended vocab; too different to merge." + }, + { + "left": "issue:43844", + "right": "issue:44936", + "accept": false, + "reason": "Gradient explosion during training and `trainer.evaluate()` failing after `trainer.train()` are not the same failure mode." + }, + { + "left": "issue:43295", + "right": "issue:43873", + "accept": false, + "reason": "Processor/tokenizer image handling and quantization offloading are unrelated subsystems." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 24, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:28218", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:42490", + "issue:42915", + "issue:43262", + "issue:43278", + "issue:43334", + "issue:43335", + "issue:43388", + "issue:43425", + "issue:43450", + "issue:43493", + "issue:43525", + "issue:43526", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43653", + "issue:43819", + "issue:43824", + "issue:43825", + "issue:43827", + "issue:43856", + "issue:43901", + "issue:43975", + "issue:44112", + "issue:44206", + "issue:44442", + "issue:44448", + "issue:44514", + "issue:44521", + "issue:44610", + "issue:44743", + "issue:44792", + "issue:44857", + "issue:45083", + "issue:45127", + "issue:45198", + "issue:45245", + "issue:45290", + "issue:45381", + "issue:45412", + "issue:45520", + "issue:45563", + "issue:45584", + "issue:45593" + ], + "soft_pairs": [ + "issue:43334|issue:43525", + "issue:44792|issue:45563", + "issue:43824|issue:43975", + "issue:43388|issue:43526", + "issue:42915|issue:43824", + "issue:43975|issue:45520", + "issue:42490|issue:43493", + "issue:28218|issue:35141", + "issue:43278|issue:43856", + "issue:38175|issue:43493", + "issue:43450|issue:43575", + "issue:43653|issue:44112", + "issue:38175|issue:45412", + "issue:44442|issue:45198", + "issue:44514|issue:44521", + "issue:44206|issue:45584", + "issue:43827|issue:43901", + "issue:43262|issue:43450", + "issue:43819|issue:44112", + "issue:44610|issue:44857", + "issue:43819|issue:45083", + "issue:37428|issue:43425", + "issue:45127|issue:45245", + "issue:43335|issue:43550", + "issue:43606|issue:45083", + "issue:44857|issue:45593", + "issue:45290|issue:45381", + "issue:44448|issue:44743", + "issue:43278|issue:43825" + ], + "prepared_review_unit_hash": "822e6bbf16e338a2c1e5a88b2f3efb4b099947d4b778664de1cf8c87542e85e6", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36537, + "estimated_input_tokens": 9135, + "estimated_eval_tokens": 18526 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft links are mostly superficial title similarity across unrelated bugs. I would not merge any of the proposed pairs; they span different models, different APIs, or different failure modes. The best issue representative is the open multimodal chat-template bug because it is user-facing, active, and has the strongest issue-level signal among this noisy set.", + "confidence": 0.84, + "canonical_issue_reason": "issue:44521 is the most representative issue here: it is open, has the strongest inbound-reference/discussion signal, and describes a concrete user-facing bug in a core multimodal API (`apply_chat_template`).", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the best standalone issue candidate for the same reasons: it is active, specific, and broader/more actionable than the other candidates in this set.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43334", + "right": "issue:43525", + "accept": false, + "reason": "Both mention missing `pad_token_id`, but they affect different model configs (`Qwen3VLTextConfig` vs `Llama4Config`) and do not clearly describe the same concrete code-path bug." + }, + { + "left": "issue:44792", + "right": "issue:45563", + "accept": false, + "reason": "These are unrelated failures: a Janus image-generation test failure versus a stale `num_return_sequences` warning in paged generation." + }, + { + "left": "issue:43824", + "right": "issue:43975", + "accept": false, + "reason": "One is an import error for a missing top-level class; the other is a detokenization regression for a specific model. Different bugs, different code paths." + }, + { + "left": "issue:43388", + "right": "issue:43526", + "accept": false, + "reason": "Both involve labels, but one is `gather_for_metrics` dropping tuple elements while the other is `BeitImageProcessorFast.reduce_labels` returning the wrong label array. Not the same bug." + }, + { + "left": "issue:42915", + "right": "issue:43824", + "accept": false, + "reason": "A Qwen3MoE FP8 config failure and a missing-export import error are unrelated." + }, + { + "left": "issue:43975", + "right": "issue:45520", + "accept": false, + "reason": "Detokenization for DeepSeek-Coder and a Python 3.13 `flash_attn` import-utils error are unrelated." + }, + { + "left": "issue:42490", + "right": "issue:43493", + "accept": false, + "reason": "A local-path import warning and a SigLIP2 correctness discrepancy are not the same underlying issue." + }, + { + "left": "issue:28218", + "right": "issue:35141", + "accept": false, + "reason": "Tokenizer spacing after added tokens and embedding reinitialization after resizing are separate tokenizer/model-init bugs." + }, + { + "left": "issue:43278", + "right": "issue:43856", + "accept": false, + "reason": "BF16/FP32 dtype drift in evaluation and Qwen3 MoE memory inefficiency are different training/runtime problems." + }, + { + "left": "issue:38175", + "right": "issue:43493", + "accept": false, + "reason": "Both mention SigLIP2, but one is zero probabilities and the other is an HF-vs-JAX discrepancy; not enough to conclude the same bug." + }, + { + "left": "issue:43450", + "right": "issue:43575", + "accept": false, + "reason": "Batched video-shape handling and tensor-parallel OOM on a large model are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44112", + "accept": false, + "reason": "A tokenizer special-token registration bug and a GraniteSpeech CI test issue are not the same defect." + }, + { + "left": "issue:38175", + "right": "issue:45412", + "accept": false, + "reason": "Both are model/runtime complaints, but one is SigLIP2 output behavior and the other is RT-DETR memory release; unrelated." + }, + { + "left": "issue:44442", + "right": "issue:45198", + "accept": false, + "reason": "AutoTokenizer loading failure and Wav2Vec2 save/tokenization failure are different tokenizer/model integration bugs." + }, + { + "left": "issue:44514", + "right": "issue:44521", + "accept": false, + "reason": "Both involve `apply_chat_template`, but one crashes on batched input with `padding=False` while the other returns all-zero assistant masks for multimodal inputs; these are not clearly the same concrete bug." + }, + { + "left": "issue:44206", + "right": "issue:45584", + "accept": false, + "reason": "An unsupported `center` argument in an audio feature extractor and Whisper failing on empty transcription are unrelated." + }, + { + "left": "issue:43827", + "right": "issue:43901", + "accept": false, + "reason": "Both are docs mismatches, but they concern different APIs and different deprecations (`pipeline()` removal vs `return_all_scores`)." + }, + { + "left": "issue:43262", + "right": "issue:43450", + "accept": false, + "reason": "Audio chat-template sampling-rate defaults and video batched-shape handling are unrelated processor bugs." + }, + { + "left": "issue:43819", + "right": "issue:44112", + "accept": false, + "reason": "A DAC latent/STE mismatch and a stale GraniteSpeech test are unrelated." + }, + { + "left": "issue:44610", + "right": "issue:44857", + "accept": false, + "reason": "Processor output resolution mismatch and an AMP crash in a loss implementation are different issues." + }, + { + "left": "issue:43819", + "right": "issue:45083", + "accept": false, + "reason": "DAC latent reconstruction mismatch and a Qwen3 Omni helper length calculation bug are unrelated." + }, + { + "left": "issue:37428", + "right": "issue:43425", + "accept": false, + "reason": "A missing flash-attention utility import and Torch 2.10 incompatibility are not the same bug." + }, + { + "left": "issue:45127", + "right": "issue:45245", + "accept": false, + "reason": "LoRA merge collapse with extended vocab and a category-count runtime limit are unrelated." + }, + { + "left": "issue:43335", + "right": "issue:43550", + "accept": false, + "reason": "A sparse-layer config bug and a torch.compile/SDPA failure in Bamba are different model/runtime problems." + }, + { + "left": "issue:43606", + "right": "issue:45083", + "accept": false, + "reason": "CPU-offload device mismatch and a Qwen3 Omni feature-length helper bug are unrelated." + }, + { + "left": "issue:44857", + "right": "issue:45593", + "accept": false, + "reason": "One is an AMP/CUDA crash in loss computation; the other is missing auxiliary losses when denoising is off. Different code paths." + }, + { + "left": "issue:45290", + "right": "issue:45381", + "accept": false, + "reason": "`apply_chat_template` tool-call handling and Qwen2.5-VL video position IDs are different multimodal bugs." + }, + { + "left": "issue:44448", + "right": "issue:44743", + "accept": false, + "reason": "Pegasus v4/v5 output differences and Qwen3 modular recurrent-state reset issues 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": 25, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:28218", + "issue:29127", + "issue:29942", + "issue:32090", + "issue:34689", + "issue:35141", + "issue:37428", + "issue:38175", + "issue:39692", + "issue:40444", + "issue:42490", + "issue:43335", + "issue:43504", + "issue:43526", + "issue:43550", + "issue:43606", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43720", + "issue:43723", + "issue:43819", + "issue:43854", + "issue:43931", + "issue:44112", + "issue:44206", + "issue:44292", + "issue:44351", + "issue:44360", + "issue:44448", + "issue:44462", + "issue:44493", + "issue:44554", + "issue:44561", + "issue:44779", + "issue:44855", + "issue:44857", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45200", + "issue:45229", + "issue:45310", + "issue:45478", + "issue:45520", + "issue:45588", + "issue:45823", + "issue:45859" + ], + "soft_pairs": [ + "issue:45200|issue:45588", + "issue:37428|issue:44351", + "issue:44206|issue:45823", + "issue:29942|issue:32090", + "issue:43611|issue:43931", + "issue:43335|issue:43653", + "issue:43550|issue:44112", + "issue:43819|issue:44493", + "issue:44855|issue:45083", + "issue:44360|issue:44561", + "issue:43335|issue:43504", + "issue:39692|issue:43526", + "issue:45310|issue:45478", + "issue:45588|issue:45859", + "issue:43335|issue:45083", + "issue:29127|issue:32090", + "issue:39692|issue:44493", + "issue:44292|issue:45520", + "issue:43720|issue:44112", + "issue:44448|issue:44779", + "issue:43335|issue:44112", + "issue:43335|issue:45072", + "issue:38175|issue:44493", + "issue:44977|issue:45229", + "issue:42490|issue:43854", + "issue:43606|issue:44857", + "issue:34689|issue:35141", + "issue:44857|issue:45072", + "issue:40444|issue:43526", + "issue:44561|issue:44779", + "issue:44448|issue:44977", + "issue:44462|issue:44561", + "issue:43723|issue:44554", + "issue:28218|issue:32090", + "issue:43701|issue:45083" + ], + "prepared_review_unit_hash": "e3e1be07932ce3d606471ff27e17f5c2767514a7241eafaf293c9b5a91cea8cb", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38156, + "estimated_input_tokens": 9539, + "estimated_eval_tokens": 19334 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a mixed bag of unrelated issues. The only clear duplicate pair is the Qwen3.5 Moe `from_pretrained` error reports; the rest share broad themes but not the same underlying bug.", + "confidence": 0.91, + "canonical_issue_reason": "Issue #45310 is the best canonical issue because it is the earlier report of the exact same Qwen3.5 Moe `from_pretrained` failure as #45478.", + "canonical_pr_reason": null, + "best_issue_reason": "#45310 is the strongest issue to keep as the representative duplicate target: same title/root problem as #45478, earlier creation, and more suitable as the primary tracker.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45200", + "right": "issue:45588", + "accept": false, + "reason": "Different bugs: Gemma 4 text-only finetuning `mm_token_type_ids` defaulting vs a Flash Attention crash on `s_aux=None`." + }, + { + "left": "issue:37428", + "right": "issue:44351", + "accept": false, + "reason": "Both are import/load errors, but they involve different missing symbols and different code paths." + }, + { + "left": "issue:44206", + "right": "issue:45823", + "accept": false, + "reason": "Both are device/runtime mismatches, but for different Gemma/Qwen model paths and different failure mechanisms." + }, + { + "left": "issue:29942", + "right": "issue:32090", + "accept": false, + "reason": "Flash Attention 2 test failures and `_gpu_broadcast_one` NoneType broadcast errors are unrelated." + }, + { + "left": "issue:43611", + "right": "issue:43931", + "accept": false, + "reason": "Both are model loading issues, but one is about `base_model_prefix` handling and the other about weight shape mismatch." + }, + { + "left": "issue:43335", + "right": "issue:43653", + "accept": false, + "reason": "Different models and failures: SwitchTransformers config vs BigBird tokenizer special-token registration." + }, + { + "left": "issue:43550", + "right": "issue:44112", + "accept": false, + "reason": "Both mention CI/device issues, but they affect different models and do not share a concrete root cause." + }, + { + "left": "issue:43819", + "right": "issue:44493", + "accept": false, + "reason": "Different problems in DAC vs positional-key warnings; same broad audio/model area only." + }, + { + "left": "issue:44855", + "right": "issue:45083", + "accept": false, + "reason": "Python 3.13 import parsing error vs a helper-function behavior bug in qwen3_omni_moe." + }, + { + "left": "issue:44360", + "right": "issue:44561", + "accept": false, + "reason": "DSA ReLU discussion is unrelated to the v5 removal of `is_torch_fx_available` breaking `trust_remote_code` models." + }, + { + "left": "issue:43335", + "right": "issue:43504", + "accept": false, + "reason": "SwitchTransformers config bug and BEiT semantic segmentation legacy-field load failure are unrelated." + }, + { + "left": "issue:39692", + "right": "issue:43526", + "accept": false, + "reason": "Docs/example errors in SigLIP2 vs a BeitImageProcessorFast `reduce_labels` bug." + }, + { + "left": "issue:45310", + "right": "issue:45478", + "accept": true, + "reason": "Exact same title and same Qwen3.5 Moe `from_pretrained` regression; #45478 is a duplicate of #45310." + }, + { + "left": "issue:45588", + "right": "issue:45859", + "accept": false, + "reason": "Different Qwen3.5-related failures: Flash Attention `s_aux=None` crash vs CPU offload incompatibility in rotary embedding." + }, + { + "left": "issue:43335", + "right": "issue:45083", + "accept": false, + "reason": "Different subsystems and symptoms: SwitchTransformers config vs qwen3_omni_moe feature-length helper behavior." + }, + { + "left": "issue:29127", + "right": "issue:32090", + "accept": false, + "reason": "LayoutLMv3 error messaging and Trainer `_gpu_broadcast_one` NoneType failures are not the same bug." + }, + { + "left": "issue:39692", + "right": "issue:44493", + "accept": false, + "reason": "SigLIP2 docs/example issues do not match the positional-id key warnings." + }, + { + "left": "issue:44292", + "right": "issue:45520", + "accept": false, + "reason": "Qwen-3-8B-NVFP4 runtime failure and `flash_attn` import KeyError are different problems." + }, + { + "left": "issue:43720", + "right": "issue:44112", + "accept": false, + "reason": "BitNet packed-weight loading and GraniteSpeech stale device override are unrelated CI/runtime issues." + }, + { + "left": "issue:44448", + "right": "issue:44779", + "accept": false, + "reason": "Pegasus v4/v5 output differences and DeepSeek tokenizer regressions are different code paths." + }, + { + "left": "issue:43335", + "right": "issue:44112", + "accept": false, + "reason": "SwitchTransformers config bug and GraniteSpeech CI failure do not share the same underlying defect." + }, + { + "left": "issue:43335", + "right": "issue:45072", + "accept": false, + "reason": "SwitchTransformers config and dtype mismatches in bfloat16 inference are unrelated." + }, + { + "left": "issue:38175", + "right": "issue:44493", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 and position-id key warnings are not the same bug." + }, + { + "left": "issue:44977", + "right": "issue:45229", + "accept": false, + "reason": "Flash-attention generation bug for Qwen3.5 is unrelated to Gemma4 multi-GPU CUDA OOM." + }, + { + "left": "issue:42490", + "right": "issue:43854", + "accept": false, + "reason": "Local-path slash import warning and GLM-4.7-Flash test loading failure are different issues." + }, + { + "left": "issue:43606", + "right": "issue:44857", + "accept": false, + "reason": "CPU offload device mismatch in Bark-small and float16 AMP crash in LwDetrImageLoss are unrelated." + }, + { + "left": "issue:34689", + "right": "issue:35141", + "accept": false, + "reason": "Llama 3.2 Vision model-loading breakage and embedding reinit on resize are different root causes." + }, + { + "left": "issue:44857", + "right": "issue:45072", + "accept": false, + "reason": "AMP/CUDA loss crash and bfloat16 dtype mismatch are separate numeric-stability issues." + }, + { + "left": "issue:40444", + "right": "issue:43526", + "accept": false, + "reason": "IterableDataset multi-image finetuning failure and BeitImageProcessorFast `reduce_labels` bug are unrelated." + }, + { + "left": "issue:44561", + "right": "issue:44779", + "accept": false, + "reason": "Removal of `is_torch_fx_available` breaking remote-code models is unrelated to DeepSeek tokenizer output regression." + }, + { + "left": "issue:44448", + "right": "issue:44977", + "accept": false, + "reason": "Different model families and symptoms: Pegasus output divergence vs Qwen3.5 flash-attention generation failure." + }, + { + "left": "issue:44462", + "right": "issue:44561", + "accept": false, + "reason": "AutoTokenizer ignoring `tokenizer.json` and v5 removal of `is_torch_fx_available` are unrelated loader regressions." + }, + { + "left": "issue:43723", + "right": "issue:44554", + "accept": false, + "reason": "Tokenizer loading in v5 and an MPS attention correctness issue are different bugs." + }, + { + "left": "issue:28218", + "right": "issue:32090", + "accept": false, + "reason": "Tokenizer adds an extra space after added tokens; unrelated to Trainer broadcast errors." + }, + { + "left": "issue:43701", + "right": "issue:45083", + "accept": false, + "reason": "Checkpoint key mismatch and qwen3_omni_moe feature-length behavior are different issues." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The duplicate target choice is grounded: #45310 and #45478 have identical titles and the earlier report is a sensible canonical issue. The summary stays conservative by describing the rest as broad-theme overlaps rather than asserting they are duplicates. The soft-edge rejections are generally title-grounded and do not overstate evidence. No PR-pair mergeability concerns apply 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": 26, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:10000", + "issue:15354", + "issue:25251", + "issue:29942", + "issue:32090", + "issue:33357", + "issue:35141", + "issue:37428", + "issue:41720", + "issue:42490", + "issue:43065", + "issue:43278", + "issue:43329", + "issue:43381", + "issue:43412", + "issue:43550", + "issue:43575", + "issue:43653", + "issue:43873", + "issue:43927", + "issue:43976", + "issue:43994", + "issue:44186", + "issue:44206", + "issue:44448", + "issue:44466", + "issue:44479", + "issue:44568", + "issue:44743", + "issue:44857", + "issue:44977", + "issue:44987", + "issue:45083", + "issue:45198", + "issue:45229", + "issue:45245", + "issue:45372", + "issue:45406", + "issue:45412", + "issue:45440", + "issue:45459", + "issue:45491", + "issue:45563", + "issue:45588", + "issue:45698", + "issue:45712", + "issue:45803", + "issue:45823" + ], + "soft_pairs": [ + "issue:37428|issue:44987", + "issue:43550|issue:45083", + "issue:43927|issue:45698", + "issue:32090|issue:33357", + "issue:29942|issue:35141", + "issue:44568|issue:45491", + "issue:43873|issue:45229", + "issue:44977|issue:45563", + "issue:41720|issue:45229", + "issue:45372|issue:45588", + "issue:43412|issue:45588", + "issue:43278|issue:43329", + "issue:45198|issue:45803", + "issue:10000|issue:15354", + "issue:42490|issue:43575", + "issue:43976|issue:45440", + "issue:45459|issue:45712", + "issue:45372|issue:45406", + "issue:44857|issue:45588", + "issue:45245|issue:45563", + "issue:44448|issue:44466", + "issue:45588|issue:45823", + "issue:43278|issue:43381", + "issue:43653|issue:44743", + "issue:43065|issue:43994", + "issue:45372|issue:45459", + "issue:44206|issue:44479", + "issue:43065|issue:45412", + "issue:25251|issue:32090", + "issue:44186|issue:45198" + ], + "prepared_review_unit_hash": "881fb2e920dc8370de6271e9cc04bbce392a311f16ccb295a303bf3542b223a6", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36994, + "estimated_input_tokens": 9249, + "estimated_eval_tokens": 18754 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is a loose collection of unrelated bugs. The soft pairs mostly share only broad themes like model loading, generation, or save/load behavior, but they do not look like the same concrete underlying issue, so none should be merged as duplicates.", + "confidence": 0.74, + "canonical_issue_reason": "issue:45588 is the most central concrete runtime bug in the cluster, and it participates in multiple soft links, making it the best representative issue here.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45588 is the strongest issue anchor: it is a specific reproducible crash, has the most cluster connectivity, and is more representative than the more niche environment/model-specific reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37428", + "right": "issue:44987", + "accept": false, + "reason": "Different failures: a missing internal flash-attention import vs a model-loading regression for a specific repo." + }, + { + "left": "issue:43550", + "right": "issue:45083", + "accept": false, + "reason": "Different models and code paths: torch.compile/SDPA on Bamba vs a Qwen3 Omni feature-length helper bug." + }, + { + "left": "issue:43927", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save/load, but one loses custom token IDs in config while the other resolves the wrong custom module." + }, + { + "left": "issue:32090", + "right": "issue:33357", + "accept": false, + "reason": "Different subsystems and symptoms: Trainer broadcast NoneType vs a MacOS bus error loading a CLIP model." + }, + { + "left": "issue:29942", + "right": "issue:35141", + "accept": false, + "reason": "Unrelated bugs: Flash Attention 2 test failures vs token embedding resize reinitializing the output embedding." + }, + { + "left": "issue:44568", + "right": "issue:45491", + "accept": false, + "reason": "Tokenizer special-token handling is unrelated to Gemma3 NaN embeddings from sliding-window attention." + }, + { + "left": "issue:43873", + "right": "issue:45229", + "accept": false, + "reason": "Both are memory-related at a high level, but one is quantization/offloading behavior and the other is a multi-GPU OOM." + }, + { + "left": "issue:44977", + "right": "issue:45563", + "accept": false, + "reason": "Different issues: a flash-attention generation regression vs a stale warning in paged generate()." + }, + { + "left": "issue:41720", + "right": "issue:45229", + "accept": false, + "reason": "Different concrete bugs: Qwen3 auto device mapping hits cudaErrorAssert, while Gemma4 31B hits an OOM." + }, + { + "left": "issue:45372", + "right": "issue:45588", + "accept": false, + "reason": "Different failure points: Gemma4 processor import breaks on ReasoningEffort, while flash_attention.py crashes on s_aux=None." + }, + { + "left": "issue:43412", + "right": "issue:45588", + "accept": false, + "reason": "Export-time executorch/erfinv issues are unrelated to a runtime flash-attention AttributeError." + }, + { + "left": "issue:43278", + "right": "issue:43329", + "accept": false, + "reason": "Embedding dtype drift in eval is unrelated to undefined video-branch helpers in multimodal token counting." + }, + { + "left": "issue:45198", + "right": "issue:45803", + "accept": false, + "reason": "Different code paths: wav2vec2 save/tokenization failure vs a bare-except bug in Fuyu tensor conversion." + }, + { + "left": "issue:10000", + "right": "issue:15354", + "accept": false, + "reason": "Completely different models and errors: DistilBert issue vs torch.jit GeneratorExp export limitation." + }, + { + "left": "issue:42490", + "right": "issue:43575", + "accept": false, + "reason": "A trailing-slash import warning is unrelated to a tensor-parallel OOM on Qwen2-57B." + }, + { + "left": "issue:43976", + "right": "issue:45440", + "accept": false, + "reason": "Environment compatibility and model-implementation divergence are different bugs." + }, + { + "left": "issue:45459", + "right": "issue:45712", + "accept": false, + "reason": "Both touch import fallbacks, but one hides tokenizer errors while the other is leftover dummy-class cleanup." + }, + { + "left": "issue:45372", + "right": "issue:45406", + "accept": false, + "reason": "Same model family, but distinct bugs: missing ReasoningEffort import vs Gemma4Processor lacking _tokenizer in serve." + }, + { + "left": "issue:44857", + "right": "issue:45588", + "accept": false, + "reason": "Different components: LwDetrImageLoss AMP crash vs flash_attention s_aux=None AttributeError." + }, + { + "left": "issue:45245", + "right": "issue:45563", + "accept": false, + "reason": "No overlap beyond being generation-related; category limit error and stale warning are unrelated." + }, + { + "left": "issue:44448", + "right": "issue:44466", + "accept": false, + "reason": "Both are v5 regressions, but one is output divergence and the other is tied-weight serialization behavior." + }, + { + "left": "issue:45588", + "right": "issue:45823", + "accept": false, + "reason": "Gemma4 flash-attention crash and Gemma4 PLE device mismatch are separate forward-path bugs." + }, + { + "left": "issue:43278", + "right": "issue:43381", + "accept": false, + "reason": "Eval-mode gradient-checkpointing restriction is unrelated to the BF16-to-FP32 embedding dtype change." + }, + { + "left": "issue:43653", + "right": "issue:44743", + "accept": false, + "reason": "Tokenizer special-token registration and recurrent-state reset are unrelated bugs." + }, + { + "left": "issue:43065", + "right": "issue:43994", + "accept": false, + "reason": "A dummy Conv2d placeholder in Sam3PixelDecoder is not the same as Siglip2 producing bad outputs." + }, + { + "left": "issue:45372", + "right": "issue:45459", + "accept": false, + "reason": "Distinct import-path bugs: missing ReasoningEffort vs tokenizer errors being swallowed by a broad exception." + }, + { + "left": "issue:44206", + "right": "issue:44479", + "accept": false, + "reason": "Different regressions: unsupported center arg in LasrFeatureExtractor vs video-input failure in Qwen VL models." + }, + { + "left": "issue:43065", + "right": "issue:45412", + "accept": false, + "reason": "A dummy layer placeholder and a model memory-leak issue are not the same underlying bug." + }, + { + "left": "issue:25251", + "right": "issue:32090", + "accept": false, + "reason": "Pipeline output-shape change is unrelated to Trainer broadcast failures." + }, + { + "left": "issue:44186", + "right": "issue:45198", + "accept": false, + "reason": "Different tokenizers and symptoms: LayoutLMv2 NER/padding crashes vs wav2vec2 save/tokenization failure." + } + ] + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:29127", + "issue:29942", + "issue:30064", + "issue:32090", + "issue:33290", + "issue:33357", + "issue:34567", + "issue:36010", + "issue:39692", + "issue:42490", + "issue:42915", + "issue:42994", + "issue:43066", + "issue:43257", + "issue:43278", + "issue:43381", + "issue:43421", + "issue:43526", + "issue:43611", + "issue:43653", + "issue:43701", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:43873", + "issue:43994", + "issue:44186", + "issue:44265", + "issue:44279", + "issue:44361", + "issue:44442", + "issue:44451", + "issue:44462", + "issue:44610", + "issue:44871", + "issue:44977", + "issue:45103", + "issue:45198", + "issue:45372", + "issue:45412", + "issue:45440", + "issue:45491", + "issue:45520", + "issue:45593", + "issue:45834", + "issue:45859", + "issue:46121" + ], + "soft_pairs": [ + "issue:44451|issue:45520", + "issue:43066|issue:44462", + "issue:44871|issue:44977", + "issue:44361|issue:45198", + "issue:43611|issue:46121", + "issue:42915|issue:42994", + "issue:43994|issue:45412", + "issue:43756|issue:45440", + "issue:33357|issue:39692", + "issue:29127|issue:34567", + "issue:32090|issue:33290", + "issue:42490|issue:43873", + "issue:29942|issue:30064", + "issue:43257|issue:43278", + "issue:43421|issue:45491", + "issue:44265|issue:45198", + "issue:44610|issue:45412", + "issue:45440|issue:45593", + "issue:43653|issue:44442", + "issue:43824|issue:44279", + "issue:43381|issue:43701", + "issue:43526|issue:45834", + "issue:43526|issue:44186", + "issue:29127|issue:33357", + "issue:43761|issue:45859", + "issue:29942|issue:36010", + "issue:45103|issue:45372" + ], + "prepared_review_unit_hash": "c46dbc793e9575e2c45d5f6bf554438193d7249926ae4517512a7e52da60c046", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 27, + "serialized_chars": 35145, + "estimated_input_tokens": 8787, + "estimated_eval_tokens": 17830 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: the soft pairs mostly share broad Transformer-area vocabulary (tokenizers, loading, quantization, vision, trainer) but not the same concrete bug or code path. I would not merge any of the soft pairs as duplicates. No PRs are present.", + "confidence": 0.81, + "canonical_issue_reason": "No true canonical duplicate stands out; if one must be chosen, issue 43873 is the broadest open runtime/loading/quantization/offloading report with the most cluster-like scope, but it does not actually subsume the others.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43873 is the best representative only in the sense of being the broadest, active infra-style bug in the set; however, the surrounding issues are too diverse for a strong canonical fit.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44451", + "right": "issue:45520", + "accept": false, + "reason": "Both are import/load failures, but one is model loading and the other is a Python 3.13 flash_attn import key error; different causes and fixes." + }, + { + "left": "issue:43066", + "right": "issue:44462", + "accept": false, + "reason": "Both concern tokenizers, but one is an eos/decoder-type mismatch and the other is repository tokenizer.json selection; not the same bug." + }, + { + "left": "issue:44871", + "right": "issue:44977", + "accept": false, + "reason": "Different models and failure modes: Gemma EOS config mismatch vs Qwen flash-attention generation issues." + }, + { + "left": "issue:44361", + "right": "issue:45198", + "accept": false, + "reason": "Unrelated tokenizer/task failures affecting different model families and code paths." + }, + { + "left": "issue:43611", + "right": "issue:46121", + "accept": false, + "reason": "One is a model-loading base_model_prefix regression; the other is a rope-validation helper TypeError. No shared concrete defect." + }, + { + "left": "issue:42915", + "right": "issue:42994", + "accept": false, + "reason": "Both touch quantization-adjacent behavior, but one is MoE weight conversion under accelerate/deepspeed and the other is quantized model saving." + }, + { + "left": "issue:43994", + "right": "issue:45412", + "accept": false, + "reason": "Nonsensical outputs in SigLIP2 vs model memory not being released in RT-DETR are unrelated bugs." + }, + { + "left": "issue:43756", + "right": "issue:45440", + "accept": false, + "reason": "Different model implementations and different structural divergences; not the same underlying issue." + }, + { + "left": "issue:33357", + "right": "issue:39692", + "accept": false, + "reason": "Both mention CLIP/SigLIP-family models, but one is a Mac bus error and the other is documentation/example mistakes." + }, + { + "left": "issue:29127", + "right": "issue:34567", + "accept": false, + "reason": "Different subsystems: LayoutLMv3 error messaging vs TrainerState token counting." + }, + { + "left": "issue:32090", + "right": "issue:33290", + "accept": false, + "reason": "Both involve Trainer/DeepSpeed contexts, but one is a NoneType broadcast error and the other is an optimizer OOM." + }, + { + "left": "issue:42490", + "right": "issue:43873", + "accept": false, + "reason": "Local-path import warning and quantization/offloading behavior are distinct loading-path issues." + }, + { + "left": "issue:29942", + "right": "issue:30064", + "accept": false, + "reason": "Flash Attention test failures and void segmentation map processing are unrelated." + }, + { + "left": "issue:43257", + "right": "issue:43278", + "accept": false, + "reason": "Qwen3 MoE weight conversion under accelerate/deepspeed is not the same as BF16-to-FP32 dtype drift during eval." + }, + { + "left": "issue:43421", + "right": "issue:45491", + "accept": false, + "reason": "Runtime special-token post-processing and Gemma3 NaN embeddings from sliding-window attention are different defects." + }, + { + "left": "issue:44265", + "right": "issue:45198", + "accept": false, + "reason": "torch.export/torch_compilable_check failure is unrelated to Wav2Vec2 save_pretrained/tokenization problems." + }, + { + "left": "issue:44610", + "right": "issue:45412", + "accept": false, + "reason": "Processor input-size mismatch and RT-DETR memory reclamation are different bugs." + }, + { + "left": "issue:45440", + "right": "issue:45593", + "accept": false, + "reason": "DeepseekV3MoE implementation drift and D-FINE auxiliary-loss behavior are not the same code-path problem." + }, + { + "left": "issue:43653", + "right": "issue:44442", + "accept": false, + "reason": "Both involve tokenizers, but one is BigBird special-token registration/empty decode and the other is AutoTokenizer loading a specific tokenizer class." + }, + { + "left": "issue:43824", + "right": "issue:44279", + "accept": false, + "reason": "An import error for a missing model class is not the same as a generic dependency issue." + }, + { + "left": "issue:43381", + "right": "issue:43701", + "accept": false, + "reason": "Gradient checkpointing in eval mode and resume_from_checkpoint key mismatch are distinct trainer-state issues." + }, + { + "left": "issue:43526", + "right": "issue:45834", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and Kosmos2.5 long OCR indexing are different image-processing bugs." + }, + { + "left": "issue:43526", + "right": "issue:44186", + "accept": false, + "reason": "Different tokenizer/image-processor paths and unrelated failure modes." + }, + { + "left": "issue:29127", + "right": "issue:33357", + "accept": false, + "reason": "LayoutLMv3 error-message clarity and a CLIP bus error on macOS are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:45859", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and Qwen3_5Moe CPU-offload incompatibility are separate model-specific bugs." + }, + { + "left": "issue:29942", + "right": "issue:36010", + "accept": false, + "reason": "Flash Attention test failures do not match a GenerationMixin import regression." + }, + { + "left": "issue:45103", + "right": "issue:45372", + "accept": false, + "reason": "Auto-docstring crashes under future annotations and Gemma 4 processor loading failing on a missing mistral_common symbol 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": 28, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:29127", + "issue:33290", + "issue:34567", + "issue:35141", + "issue:42490", + "issue:42915", + "issue:43232", + "issue:43262", + "issue:43316", + "issue:43334", + "issue:43381", + "issue:43388", + "issue:43441", + "issue:43525", + "issue:43526", + "issue:43531", + "issue:43575", + "issue:43646", + "issue:43701", + "issue:43723", + "issue:43761", + "issue:43931", + "issue:44062", + "issue:44077", + "issue:44079", + "issue:44188", + "issue:44206", + "issue:44387", + "issue:44442", + "issue:44464", + "issue:44479", + "issue:44625", + "issue:44743", + "issue:44811", + "issue:44857", + "issue:44898", + "issue:44977", + "issue:45003", + "issue:45072", + "issue:45245", + "issue:45290", + "issue:45440", + "issue:45479", + "issue:45491", + "issue:45584", + "issue:45588", + "issue:45803", + "issue:45907" + ], + "soft_pairs": [ + "issue:43334|issue:43531", + "issue:44625|issue:45584", + "issue:29127|issue:35141", + "issue:44977|issue:45245", + "issue:42915|issue:43931", + "issue:44442|issue:45072", + "issue:43526|issue:44857", + "issue:44464|issue:45491", + "issue:43262|issue:44206", + "issue:42490|issue:43381", + "issue:43575|issue:43701", + "issue:43575|issue:44387", + "issue:44062|issue:45907", + "issue:44743|issue:44811", + "issue:43441|issue:45588", + "issue:43646|issue:44077", + "issue:43761|issue:44079", + "issue:45003|issue:45440", + "issue:43232|issue:43388", + "issue:43723|issue:45479", + "issue:44479|issue:45290", + "issue:43316|issue:43525", + "issue:44442|issue:45803", + "issue:43761|issue:44188", + "issue:44442|issue:44898", + "issue:42490|issue:44743", + "issue:33290|issue:34567" + ], + "prepared_review_unit_hash": "64428ce8bd1ca603d69cd72db9eeeccefa9462b340a01c766ff6585060cd8629", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 27, + "serialized_chars": 35905, + "estimated_input_tokens": 8977, + "estimated_eval_tokens": 18210 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No duplicate cluster here: the items are a set of unrelated bug reports spanning configs, tokenizers, generation, vision/audio processors, attention kernels, and memory/runtime issues. All soft pairs should be rejected.", + "confidence": 0.97, + "canonical_issue_reason": "No single canonical issue; the items are unrelated regressions across different subsystems, so none is a safe duplicate anchor.", + "canonical_pr_reason": null, + "best_issue_reason": "No strong cluster representative for the same reason: the reports do not describe one underlying bug or change.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43334", + "right": "issue:43531", + "accept": false, + "reason": "Different bugs: Qwen3-VL pad_token_id AttributeError vs Qwen3-MoE sliding_window behavior." + }, + { + "left": "issue:44625", + "right": "issue:45584", + "accept": false, + "reason": "Unrelated: Qwen3.5 num_labels propagation vs Whisper empty-transcription/chat-template failure." + }, + { + "left": "issue:29127", + "right": "issue:35141", + "accept": false, + "reason": "LayoutLMv3 error messaging is unrelated to token embedding resize/post_init reinitialization." + }, + { + "left": "issue:44977", + "right": "issue:45245", + "accept": false, + "reason": "Different failure modes: Qwen3.5 flash-attention generation vs category-count runtime limit." + }, + { + "left": "issue:42915", + "right": "issue:43931", + "accept": false, + "reason": "Qwen3Moe FP8 config failure and Qwen3-VL weight-shape mismatch are separate model-loading bugs." + }, + { + "left": "issue:44442", + "right": "issue:45072", + "accept": false, + "reason": "Tokenizer loading failure is unrelated to dtype mismatches in inference wrappers." + }, + { + "left": "issue:43526", + "right": "issue:44857", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is unrelated to LwDetrImageLoss AMP/CUDA crash." + }, + { + "left": "issue:44464", + "right": "issue:45491", + "accept": false, + "reason": "Compiled forward chunked-generation inconsistency is not the same as Gemma3 sliding-window NaNs." + }, + { + "left": "issue:43262", + "right": "issue:44206", + "accept": false, + "reason": "Audio processor sampling-rate default mismatch vs unsupported center arg in LasrFeatureExtractor." + }, + { + "left": "issue:42490", + "right": "issue:43381", + "accept": false, + "reason": "Local-path trailing-slash warning is unrelated to gradient-checkpointing eval-mode behavior." + }, + { + "left": "issue:43575", + "right": "issue:43701", + "accept": false, + "reason": "TP OOM on Qwen2-57B and resume_from_checkpoint key mismatch are different problems." + }, + { + "left": "issue:43575", + "right": "issue:44387", + "accept": false, + "reason": "Both mention OOM, but one is tensor parallel loading and the other is int4 reserved-memory regression." + }, + { + "left": "issue:44062", + "right": "issue:45907", + "accept": false, + "reason": "AddedToken kwarg collision is unrelated to list_repo_files not forwarding revision." + }, + { + "left": "issue:44743", + "right": "issue:44811", + "accept": false, + "reason": "Recurrent-state reset in Qwen3.5 is unrelated to Whisper batch_decode skip_special_tokens handling." + }, + { + "left": "issue:43441", + "right": "issue:45588", + "accept": false, + "reason": "Ministral-3 FlashAttention failure and sink-less model s_aux=None crash are different code paths." + }, + { + "left": "issue:43646", + "right": "issue:44077", + "accept": false, + "reason": "Broad custom-model-init breakage is not the same as patchtsmixer post_init allowance." + }, + { + "left": "issue:43761", + "right": "issue:44079", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and ModelOutput key assignment bug are unrelated." + }, + { + "left": "issue:45003", + "right": "issue:45440", + "accept": false, + "reason": "Unsafe sys.modules access is unrelated to DeepseekV3MoE implementation divergence." + }, + { + "left": "issue:43232", + "right": "issue:43388", + "accept": false, + "reason": "Generation kwargs update after sync_gpus and metric-gather label truncation are separate bugs." + }, + { + "left": "issue:43723", + "right": "issue:45479", + "accept": false, + "reason": "Tokenizer loading regression and single-label-classification zero-loss bug do not share a code path." + }, + { + "left": "issue:44479", + "right": "issue:45290", + "accept": false, + "reason": "Video-input regression in Qwen VL models is unrelated to chat-template tokenization with tool calls." + }, + { + "left": "issue:43316", + "right": "issue:43525", + "accept": false, + "reason": "API discrepancy in Gemma3TextConfig is not the same as missing pad_token_id in Llama4Config." + }, + { + "left": "issue:44442", + "right": "issue:45803", + "accept": false, + "reason": "AutoTokenizer failure and bare-except in FuyuBatchFeature are unrelated." + }, + { + "left": "issue:43761", + "right": "issue:44188", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and torch.compile attention-kernel divergence are different issues." + }, + { + "left": "issue:44442", + "right": "issue:44898", + "accept": false, + "reason": "Tokenizer loading bug is unrelated to Perceiver non-default-resolution interpolation failure." + }, + { + "left": "issue:42490", + "right": "issue:44743", + "accept": false, + "reason": "Import warning on trailing slash path and Qwen3.5 recurrent-state reset are unrelated." + }, + { + "left": "issue:33290", + "right": "issue:34567", + "accept": false, + "reason": "AdaFactor/DeepSpeed OOM and TrainerState num_input_tokens_seen not updating 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": 29, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:16998", + "issue:22355", + "issue:25251", + "issue:28282", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:36010", + "issue:36331", + "issue:41628", + "issue:42175", + "issue:42907", + "issue:43065", + "issue:43295", + "issue:43316", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43493", + "issue:43540", + "issue:43611", + "issue:43643", + "issue:43673", + "issue:43720", + "issue:43749", + "issue:43825", + "issue:43827", + "issue:43873", + "issue:43931", + "issue:44077", + "issue:44162", + "issue:44190", + "issue:44442", + "issue:44479", + "issue:44496", + "issue:44514", + "issue:44617", + "issue:44743", + "issue:44792", + "issue:44855", + "issue:44871", + "issue:44877", + "issue:44913", + "issue:45137", + "issue:45216", + "issue:45372", + "issue:45698", + "issue:45759" + ], + "soft_pairs": [ + "issue:33290|issue:34689", + "issue:33290|issue:35141", + "issue:43381|issue:43873", + "issue:34689|issue:43611", + "issue:42907|issue:45698", + "issue:44077|issue:44479", + "issue:43931|issue:45216", + "issue:43720|issue:44442", + "issue:43493|issue:43827", + "issue:43316|issue:45759", + "issue:41628|issue:45372", + "issue:34689|issue:36010", + "issue:33290|issue:36331", + "issue:43749|issue:43931", + "issue:43873|issue:44871", + "issue:43643|issue:44913", + "issue:44190|issue:44496", + "issue:44855|issue:45137", + "issue:43540|issue:44514", + "issue:43316|issue:44877", + "issue:43065|issue:44617", + "issue:43404|issue:43441", + "issue:42175|issue:43316", + "issue:25251|issue:28282", + "issue:44162|issue:44792", + "issue:43295|issue:43825", + "issue:43673|issue:44743", + "issue:16998|issue:22355" + ], + "prepared_review_unit_hash": "1ab02f30bf2128d19719b5654b8ccd182439d4679d8b413322fb55027fb29256", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36095, + "estimated_input_tokens": 9024, + "estimated_eval_tokens": 18304 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft pairs are mostly superficial text similarities around loading/import/config, but they describe different concrete bugs, models, or code paths. None should be merged as duplicate pairs.", + "confidence": 0.82, + "canonical_issue_reason": "issue:43749 is the best representative issue to anchor triage because it has the broadest loading/infra scope and the strongest discussion/inbound-ref signal in this set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43749 is the highest-signal, most broadly relevant issue here, but it still does not match the other items closely enough to serve as a true duplicate canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:33290", + "right": "issue:34689", + "accept": false, + "reason": "Different failures: Adafactor/DeepSpeed OOM vs a specific Llama 3.2 vision model-loading regression." + }, + { + "left": "issue:33290", + "right": "issue:35141", + "accept": false, + "reason": "Training OOM and embedding-resize post_init reinitialization are unrelated code paths." + }, + { + "left": "issue:43381", + "right": "issue:43873", + "accept": false, + "reason": "Eval-mode gradient checkpointing issue vs quantization/offloading bug." + }, + { + "left": "issue:34689", + "right": "issue:43611", + "accept": false, + "reason": "Both are loading regressions, but one is a model-specific Llama 3.2 breakage and the other is a base_model_prefix loader regression." + }, + { + "left": "issue:42907", + "right": "issue:45698", + "accept": false, + "reason": "Both involve save/load, but one is dequantized model saving and the other is wrong custom module resolution after save_pretrained." + }, + { + "left": "issue:44077", + "right": "issue:44479", + "accept": false, + "reason": "Unrelated issues: patchtsmixer post_init validation vs Qwen video-input regression." + }, + { + "left": "issue:43931", + "right": "issue:45216", + "accept": false, + "reason": "Different models and failures: Qwen3-VL weight-shape mismatch vs Qwen3.5 checkpoint saving corruption." + }, + { + "left": "issue:43720", + "right": "issue:44442", + "accept": false, + "reason": "Packed-weight accelerate loading bug vs tokenizer import failure." + }, + { + "left": "issue:43493", + "right": "issue:43827", + "accept": false, + "reason": "SigLIP2 implementation mismatch vs stale docs referencing pipeline() after v5 removals." + }, + { + "left": "issue:43316", + "right": "issue:45759", + "accept": false, + "reason": "Different API mismatches in different model families; not the same bug." + }, + { + "left": "issue:41628", + "right": "issue:45372", + "accept": false, + "reason": "Both are import errors, but they involve different modules and different root causes." + }, + { + "left": "issue:34689", + "right": "issue:36010", + "accept": false, + "reason": "Separate loader/import regressions with different missing symbols and code paths." + }, + { + "left": "issue:33290", + "right": "issue:36331", + "accept": false, + "reason": "Optimizer OOM under DeepSpeed is unrelated to the CustomTrainer compute_loss signature break." + }, + { + "left": "issue:43749", + "right": "issue:43931", + "accept": false, + "reason": "Different loading subsystems: FSDP CPU-efficient loading vs a Qwen3-VL shape mismatch." + }, + { + "left": "issue:43873", + "right": "issue:44871", + "accept": false, + "reason": "Quantization/offloading behavior vs eos_token_id configuration inconsistency." + }, + { + "left": "issue:43643", + "right": "issue:44913", + "accept": false, + "reason": "Remote-code config deserialization issue vs GPTNeoX config field reverting on reload." + }, + { + "left": "issue:44190", + "right": "issue:44496", + "accept": false, + "reason": "Dataset loading script problem vs unrecognized model/config loading error." + }, + { + "left": "issue:44855", + "right": "issue:45137", + "accept": false, + "reason": "Python 3.13 torch.jit parsing problem vs DeepSpeed ZeRO3 deque error." + }, + { + "left": "issue:43540", + "right": "issue:44514", + "accept": false, + "reason": "Different multimodal bugs: video processing failure vs batched chat-template padding crash." + }, + { + "left": "issue:43316", + "right": "issue:44877", + "accept": false, + "reason": "Gemma3TextConfig API inconsistency vs strict config rejection for granite_speech." + }, + { + "left": "issue:43065", + "right": "issue:44617", + "accept": false, + "reason": "Dummy Conv2d design issue vs CUDA OOM in Sam3Video." + }, + { + "left": "issue:43404", + "right": "issue:43441", + "accept": false, + "reason": "Different Mistral-family bugs: lm_head weight tying vs FlashAttention regression." + }, + { + "left": "issue:42175", + "right": "issue:43316", + "accept": false, + "reason": "Package dependency/extra install issue vs model config API discrepancy." + }, + { + "left": "issue:25251", + "right": "issue:28282", + "accept": false, + "reason": "Pipeline output-format bug vs missing PyTorch dependency import error." + }, + { + "left": "issue:44162", + "right": "issue:44792", + "accept": false, + "reason": "ESM2 runtime breakage and janus test failure are unrelated." + }, + { + "left": "issue:43295", + "right": "issue:43825", + "accept": false, + "reason": "Custom processor/tokenizer regression vs pipeline() error-message/documentation issue." + }, + { + "left": "issue:43673", + "right": "issue:44743", + "accept": false, + "reason": "Cache/chunked_prefill issue vs recurrent-state reset in qwen3_5." + }, + { + "left": "issue:16998", + "right": "issue:22355", + "accept": false, + "reason": "Model_max_length question vs missing transformers.onnx module import error." + } + ] + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:32090", + "issue:35141", + "issue:41950", + "issue:42617", + "issue:43065", + "issue:43295", + "issue:43388", + "issue:43452", + "issue:43475", + "issue:43504", + "issue:43525", + "issue:43531", + "issue:43550", + "issue:43575", + "issue:43606", + "issue:43646", + "issue:43688", + "issue:43701", + "issue:43742", + "issue:43761", + "issue:43906", + "issue:43931", + "issue:44060", + "issue:44077", + "issue:44190", + "issue:44242", + "issue:44464", + "issue:44479", + "issue:44496", + "issue:44521", + "issue:44534", + "issue:44610", + "issue:44617", + "issue:44734", + "issue:44792", + "issue:44871", + "issue:44913", + "issue:44964", + "issue:44977", + "issue:45127", + "issue:45200", + "issue:45290", + "issue:45381", + "issue:45406", + "issue:45563", + "issue:45584", + "issue:45698", + "issue:46050" + ], + "soft_pairs": [ + "issue:43550|issue:43701", + "issue:44077|issue:44534", + "issue:41950|issue:43475", + "issue:43606|issue:43701", + "issue:44871|issue:45584", + "issue:43388|issue:43688", + "issue:43761|issue:44479", + "issue:42617|issue:43575", + "issue:43295|issue:44871", + "issue:43575|issue:43742", + "issue:45127|issue:45200", + "issue:43452|issue:44496", + "issue:44871|issue:46050", + "issue:43646|issue:44913", + "issue:43504|issue:45698", + "issue:44734|issue:45406", + "issue:44617|issue:45584", + "issue:32090|issue:35141", + "issue:45563|issue:45584", + "issue:43761|issue:44242", + "issue:43525|issue:43531", + "issue:43931|issue:44060", + "issue:44792|issue:44977", + "issue:44479|issue:44734", + "issue:43906|issue:45290", + "issue:44464|issue:45563", + "issue:44190|issue:44964", + "issue:44521|issue:45381", + "issue:43065|issue:44610" + ], + "prepared_review_unit_hash": "ae5f45b2d8e49743eeb9560829cf8ee5264eacce28adc252a5271721d15a8bc8", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36412, + "estimated_input_tokens": 9103, + "estimated_eval_tokens": 18462 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft pairs are mostly superficial similarity matches across unrelated Transformers bugs; none are true duplicates, and there are no PRs in this cluster.", + "confidence": 0.94, + "canonical_issue_reason": "issue:44521 is the most active open issue with the strongest downstream signal (multiple inbound references), so it is the best representative if one issue must anchor this heterogeneous cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the most suitable issue to surface for triage because it is open, well-discussed, and externally referenced, even though it is not a duplicate of the other items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43550", + "right": "issue:43701", + "accept": false, + "reason": "Different bugs: torch.compile/SDPA failure vs resume_from_checkpoint key mismatch." + }, + { + "left": "issue:44077", + "right": "issue:44534", + "accept": false, + "reason": "API/post-init policy issue vs non-persistent buffer initialization regression." + }, + { + "left": "issue:41950", + "right": "issue:43475", + "accept": false, + "reason": "Pipeline processor selection bug vs Sam3Video output attribute error; unrelated code paths." + }, + { + "left": "issue:43606", + "right": "issue:43701", + "accept": false, + "reason": "CPU offload device mismatch vs checkpoint key mismatch; not the same failure." + }, + { + "left": "issue:44871", + "right": "issue:45584", + "accept": false, + "reason": "EOS token config inconsistency vs Whisper empty-transcription crash after token alignment." + }, + { + "left": "issue:43388", + "right": "issue:43688", + "accept": false, + "reason": "Last-batch metric gathering bug vs MoE auxiliary-loss normalization bug." + }, + { + "left": "issue:43761", + "right": "issue:44479", + "accept": false, + "reason": "CLIPVision hidden_states regression vs Qwen video-input regression; different models and symptoms." + }, + { + "left": "issue:42617", + "right": "issue:43575", + "accept": false, + "reason": "Script/runtime failure vs OOM on large-model tensor-parallel loading." + }, + { + "left": "issue:43295", + "right": "issue:44871", + "accept": false, + "reason": "Gemma4 processor/tokenizer regression vs Gemma eos_token_id config mismatch." + }, + { + "left": "issue:43575", + "right": "issue:43742", + "accept": false, + "reason": "OOM on Qwen2-57B load vs MobileLLM config/key error; unrelated load failures." + }, + { + "left": "issue:45127", + "right": "issue:45200", + "accept": false, + "reason": "LoRA merge with tied embeddings vs multimodal token_type_ids defaulting; different mechanisms." + }, + { + "left": "issue:43452", + "right": "issue:44496", + "accept": false, + "reason": "GGUF tokenizer/model loading path bug vs missing model_type in config.json." + }, + { + "left": "issue:44871", + "right": "issue:46050", + "accept": false, + "reason": "EOS config mismatch vs quantized save/load breakage; not the same underlying bug." + }, + { + "left": "issue:43646", + "right": "issue:44913", + "accept": false, + "reason": "Custom model initialization regression vs GPTNeoX config reload reverting rotary_pct." + }, + { + "left": "issue:43504", + "right": "issue:45698", + "accept": false, + "reason": "Legacy-field pretrained load issue vs wrong custom module loaded after save_pretrained." + }, + { + "left": "issue:44734", + "right": "issue:45406", + "accept": false, + "reason": "KV-cache continuation indexing crash vs Gemma4Processor missing _tokenizer." + }, + { + "left": "issue:44617", + "right": "issue:45584", + "accept": false, + "reason": "Sam3Video CUDA OOM vs Whisper empty-transcription generation failure." + }, + { + "left": "issue:32090", + "right": "issue:35141", + "accept": false, + "reason": "Trainer GPU broadcast of NoneType vs output embedding reinit in post_init." + }, + { + "left": "issue:45563", + "right": "issue:45584", + "accept": false, + "reason": "Stale warning for num_return_sequences vs actual generation failure on empty transcription." + }, + { + "left": "issue:43761", + "right": "issue:44242", + "accept": false, + "reason": "Vision-model hidden_states regression vs load-balancing loss omission in MoE routing." + }, + { + "left": "issue:43525", + "right": "issue:43531", + "accept": false, + "reason": "Missing pad_token_id on Llama4Config vs Qwen3-MoE sliding_window handling." + }, + { + "left": "issue:43931", + "right": "issue:44060", + "accept": false, + "reason": "Qwen3-VL weight-shape mismatch vs incorrect tied-weights warning in Qwen3-Next." + }, + { + "left": "issue:44792", + "right": "issue:44977", + "accept": false, + "reason": "Janus image-generation test failure vs Qwen3.5 flash-attention generation bug." + }, + { + "left": "issue:44479", + "right": "issue:44734", + "accept": false, + "reason": "Video-input regression in Qwen VL models vs serve KV-cache indexing crash." + }, + { + "left": "issue:43906", + "right": "issue:45290", + "accept": false, + "reason": "Repro of an older issue vs assistant-message/tool-call crash; similar area but not clearly the same bug." + }, + { + "left": "issue:44464", + "right": "issue:45563", + "accept": false, + "reason": "Compiled-forward chunked generation inconsistency vs a stale warning about num_return_sequences." + }, + { + "left": "issue:44190", + "right": "issue:44964", + "accept": false, + "reason": "Local dataset loading script issue vs model-loading failure for Phi-4 multimodal." + }, + { + "left": "issue:44521", + "right": "issue:45381", + "accept": false, + "reason": "All-zero assistant masks in multimodal templates vs wrong vision_position_ids for Qwen video input." + }, + { + "left": "issue:44521", + "right": "issue:45290", + "accept": false, + "reason": "Multimodal assistant-mask bug vs assistant-message/tool-call crash; related surface area but different concrete failures." + }, + { + "left": "issue:44521", + "right": "issue:45584", + "accept": false, + "reason": "Multimodal chat-template masking vs Whisper empty-transcription handling; unrelated models and paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 31, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:35141", + "issue:36246", + "issue:38617", + "issue:39692", + "issue:41628", + "issue:41720", + "issue:41950", + "issue:42490", + "issue:43295", + "issue:43335", + "issue:43479", + "issue:43540", + "issue:43550", + "issue:43644", + "issue:43650", + "issue:43827", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44077", + "issue:44079", + "issue:44112", + "issue:44190", + "issue:44220", + "issue:44462", + "issue:44496", + "issue:44534", + "issue:44734", + "issue:44737", + "issue:44792", + "issue:44855", + "issue:44913", + "issue:44933", + "issue:44938", + "issue:45127", + "issue:45245", + "issue:45276", + "issue:45357", + "issue:45362", + "issue:45375", + "issue:45542", + "issue:45584", + "issue:45588", + "issue:45600", + "issue:45803", + "issue:45907", + "issue:46082", + "issue:46121" + ], + "soft_pairs": [ + "issue:43827|issue:43994", + "issue:44534|issue:44913", + "issue:41720|issue:45245", + "issue:44792|issue:45600", + "issue:44855|issue:45588", + "issue:44933|issue:45362", + "issue:35141|issue:42490", + "issue:45127|issue:45357", + "issue:43335|issue:44855", + "issue:44737|issue:45803", + "issue:44462|issue:45907", + "issue:43550|issue:44190", + "issue:43976|issue:44855", + "issue:44077|issue:44079", + "issue:44938|issue:45542", + "issue:43644|issue:44913", + "issue:43540|issue:44734", + "issue:44112|issue:45600", + "issue:43335|issue:44190", + "issue:46082|issue:46121", + "issue:39692|issue:44190", + "issue:44496|issue:45803", + "issue:36246|issue:41628", + "issue:45276|issue:45584", + "issue:43479|issue:45375", + "issue:45245|issue:45584", + "issue:43650|issue:44220", + "issue:43295|issue:45276", + "issue:41950|issue:44016", + "issue:36246|issue:38617" + ], + "prepared_review_unit_hash": "132b057c25b57e9143c110f66208f234b5d6b1425818a62302343e9c1606f74a", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36924, + "estimated_input_tokens": 9231, + "estimated_eval_tokens": 18718 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a set of unrelated closed/open issues that only share broad topical keywords; I did not find any soft pair that clearly describes the same underlying bug. The closest overlap is in token-embedding resize behavior, but even that does not rise to a safe duplicate merge.", + "confidence": 0.89, + "canonical_issue_reason": "issue:35141 is the cleanest and most general report in the only somewhat related embedding-resize thread: it pinpoints the post_init reinitialization problem when tie_word_embedding=False. The nearby embedding report (issue:45276) is more model-specific and not a confirmed duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:35141 is the best representative issue because it states a concrete, root-cause-level bug in the core resize_token_embeddings path and is more broadly actionable than the more specialized reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43827", + "right": "issue:43994", + "accept": false, + "reason": "Docs still mentioning pipeline() and a SigLIP2 inference mismatch/quantization problem are different bug classes." + }, + { + "left": "issue:44534", + "right": "issue:44913", + "accept": false, + "reason": "One is about junk-filled non-persistent buffers in v5; the other is about rotary_pct not persisting on reload." + }, + { + "left": "issue:41720", + "right": "issue:45245", + "accept": false, + "reason": "Qwen3 auto device-map CUDA asserts and a categorical limit RuntimeError are unrelated." + }, + { + "left": "issue:44792", + "right": "issue:45600", + "accept": false, + "reason": "A Janus image-generation test failure and a stale auto_mappings CI break are different code paths." + }, + { + "left": "issue:44855", + "right": "issue:45588", + "accept": false, + "reason": "Python 3.13 import parsing failure in DebertaV2 and a flash-attention sink_aux crash are unrelated." + }, + { + "left": "issue:44933", + "right": "issue:45362", + "accept": false, + "reason": "Missing image_utils import and a Qwen3.5 chat crash do not share the same underlying bug." + }, + { + "left": "issue:35141", + "right": "issue:42490", + "accept": false, + "reason": "Token-embedding resize/reinit behavior is unrelated to a local-path trailing-slash warning." + }, + { + "left": "issue:45127", + "right": "issue:45357", + "accept": false, + "reason": "LoRA merge collapse with extended vocab and incorrect visual encoder keys in save_pretrained are different failures." + }, + { + "left": "issue:43335", + "right": "issue:44855", + "accept": false, + "reason": "SwitchTransformers sparse-layer construction and a Python 3.13 import parsing error are unrelated." + }, + { + "left": "issue:44737", + "right": "issue:45803", + "accept": false, + "reason": "XLNet device placement in relative positional encoding and a bare-except bug in Fuyu are different issues." + }, + { + "left": "issue:44462", + "right": "issue:45907", + "accept": false, + "reason": "Ignoring tokenizer.json and forwarding list_repo_files revision are separate repository/API bugs." + }, + { + "left": "issue:43550", + "right": "issue:44190", + "accept": false, + "reason": "A Bamba torch.compile SDPA failure and a local dataset loading failure in a training script are unrelated." + }, + { + "left": "issue:43976", + "right": "issue:44855", + "accept": false, + "reason": "Python version compatibility breakage and a DebertaV2 import indentation/parsing issue are not the same bug." + }, + { + "left": "issue:44077", + "right": "issue:44079", + "accept": false, + "reason": "Optional post_init enforcement in patchtsmixer and ModelOutput key assignment with None are different internal behaviors." + }, + { + "left": "issue:44938", + "right": "issue:45542", + "accept": false, + "reason": "Python 3.14 load failure and missing TensorFlow backend handling are unrelated packaging/runtime problems." + }, + { + "left": "issue:43644", + "right": "issue:44913", + "accept": false, + "reason": "Non-persistent buffers filled with junk and rotary_pct resetting on reload are different serialization/config issues." + }, + { + "left": "issue:43540", + "right": "issue:44734", + "accept": false, + "reason": "Qwen3OmniMoe video-input validation and serve KV-cache tensor indexing are distinct code paths." + }, + { + "left": "issue:44112", + "right": "issue:45600", + "accept": false, + "reason": "GraniteSpeech CI flakiness from stale device overrides is unrelated to removed Sam3LiteText config references." + }, + { + "left": "issue:46082", + "right": "issue:46121", + "accept": false, + "reason": "LlamaConfig head_dim validation and rope-parameter JSON list handling are separate configuration bugs." + }, + { + "left": "issue:39692", + "right": "issue:44190", + "accept": false, + "reason": "SigLIP2 documentation errors and a dataset-loading failure in a training script are not duplicates." + }, + { + "left": "issue:44496", + "right": "issue:45803", + "accept": false, + "reason": "An unrecognized-model/config loading error and a bare-except bug are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:41628", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and modules, not the same export bug." + }, + { + "left": "issue:45276", + "right": "issue:45584", + "accept": false, + "reason": "Gemma4 embedding-resize behavior and Whisper empty-transcription generation failure are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:45375", + "accept": false, + "reason": "Default multimodal config initialization and a missing deepstack_visual_indexes field are different config problems." + }, + { + "left": "issue:45245", + "right": "issue:45584", + "accept": false, + "reason": "A category-count overflow and a Whisper empty-transcription bug are unrelated." + }, + { + "left": "issue:43650", + "right": "issue:44220", + "accept": false, + "reason": "The placeholder title 'ADD THE DATA' provides no overlap with the fbank feature extraction bug." + }, + { + "left": "issue:43295", + "right": "issue:45276", + "accept": false, + "reason": "Processor/tokenizer regression and Gemma4 resize_token_embeddings behavior concern different multimodal and embedding paths." + }, + { + "left": "issue:41950", + "right": "issue:44016", + "accept": false, + "reason": "A video-classification pipeline processor lookup issue and a notebook syntax error are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:38617", + "accept": false, + "reason": "These are separate import errors for different symbols in different modules." + } + ] + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:38175", + "issue:42175", + "issue:42490", + "issue:42907", + "issue:43065", + "issue:43295", + "issue:43299", + "issue:43335", + "issue:43381", + "issue:43404", + "issue:43441", + "issue:43526", + "issue:43531", + "issue:43643", + "issue:43673", + "issue:43824", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44062", + "issue:44075", + "issue:44112", + "issue:44186", + "issue:44220", + "issue:44246", + "issue:44279", + "issue:44336", + "issue:44368", + "issue:44496", + "issue:44561", + "issue:44683", + "issue:44734", + "issue:44933", + "issue:44964", + "issue:44987", + "issue:45003", + "issue:45042", + "issue:45216", + "issue:45335", + "issue:45357", + "issue:45446", + "issue:45584", + "issue:45593", + "issue:45600", + "issue:45632", + "issue:45759" + ], + "soft_pairs": [ + "issue:44186|issue:44734", + "issue:42175|issue:44279", + "issue:44683|issue:45446", + "issue:38175|issue:43065", + "issue:44246|issue:45003", + "issue:44016|issue:44220", + "issue:43404|issue:45216", + "issue:44734|issue:45632", + "issue:44561|issue:45632", + "issue:43643|issue:44496", + "issue:44964|issue:45600", + "issue:44112|issue:44964", + "issue:43854|issue:44964", + "issue:43976|issue:44987", + "issue:42907|issue:43441", + "issue:42490|issue:45632", + "issue:43531|issue:45600", + "issue:43673|issue:44336", + "issue:43295|issue:45335", + "issue:44016|issue:44062", + "issue:43526|issue:43901", + "issue:43824|issue:44933", + "issue:44496|issue:45042", + "issue:43335|issue:45593", + "issue:43299|issue:45216", + "issue:44075|issue:44368", + "issue:43976|issue:44279", + "issue:43381|issue:45593", + "issue:43901|issue:44496", + "issue:45335|issue:45584", + "issue:44496|issue:45759", + "issue:43299|issue:45357" + ], + "prepared_review_unit_hash": "6d0803fc46f7affd8d77bbfe73d9b0bb9d64a0818ca8440f9f1f6b8f358dc1ad", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 32, + "serialized_chars": 36653, + "estimated_input_tokens": 9164, + "estimated_eval_tokens": 18584 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is heterogeneous: the items cover unrelated regressions across model loading, tokenizers, save/load behavior, attention, docs, and CI. None of the soft-edge pairs look like the same concrete bug/change, so I would not merge any of them as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": "No single canonical issue stands out because the set spans many different failure modes and subsystems rather than one underlying bug.", + "canonical_pr_reason": null, + "best_issue_reason": "No strong global best issue; if forced, a concrete repro-oriented model-loading report would be the most representative, but the cluster is too mixed for a true canonical.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44186", + "right": "issue:44734", + "accept": false, + "reason": "Different code paths: LayoutLMv2 NER/padding crash versus serve KV-cache continuation indexing." + }, + { + "left": "issue:42175", + "right": "issue:44279", + "accept": false, + "reason": "One is a missing TensorFlow backend in installation extras; the other is a vague dependency problem." + }, + { + "left": "issue:44683", + "right": "issue:45446", + "accept": false, + "reason": "Both mention flex_attention, but one is a torch>=2.9 compilation failure and the other is a version-check/import bug for AuxRequest." + }, + { + "left": "issue:38175", + "right": "issue:43065", + "accept": false, + "reason": "Unrelated models and failures: SigLIP2 zero probabilities versus a dummy Conv2d in Sam3PixelDecoder." + }, + { + "left": "issue:44246", + "right": "issue:45003", + "accept": false, + "reason": "Import slowness and unsafe sys.modules access may be adjacent, but they are not the same concrete bug." + }, + { + "left": "issue:44016", + "right": "issue:44220", + "accept": false, + "reason": "A notebook syntax error is unrelated to the _torch_extract_fbank_features issue." + }, + { + "left": "issue:43404", + "right": "issue:45216", + "accept": false, + "reason": "Different bugs: untied lm_head weights in Mistral3 versus incorrect save_pretrained output for Qwen3.5." + }, + { + "left": "issue:44734", + "right": "issue:45632", + "accept": false, + "reason": "Serve KV-cache continuation crash is unrelated to local-model cache path collisions with trust_remote_code." + }, + { + "left": "issue:44561", + "right": "issue:45632", + "accept": false, + "reason": "Removal of is_torch_fx_available affecting trust_remote_code is a different regression from local-model cache path collisions." + }, + { + "left": "issue:43643", + "right": "issue:44496", + "accept": false, + "reason": "Missing fields from trust_remote_code AutoConfig and an unrecognized-model/model_type error are different loading failures." + }, + { + "left": "issue:44964", + "right": "issue:45600", + "accept": false, + "reason": "Phi-4 multimodal loading failure is unrelated to stale auto_mappings references to Sam3LiteText configs." + }, + { + "left": "issue:44112", + "right": "issue:44964", + "accept": false, + "reason": "CI stale device override test failure is not the same as model loading for Phi-4 multimodal." + }, + { + "left": "issue:43854", + "right": "issue:44964", + "accept": false, + "reason": "Different model families and failure modes: GLM-4.7-Flash unit-test loading versus Phi-4 multimodal loading." + }, + { + "left": "issue:43976", + "right": "issue:44987", + "accept": false, + "reason": "Python version compatibility problem is unrelated to loading physical-intelligence/fast." + }, + { + "left": "issue:42907", + "right": "issue:43441", + "accept": false, + "reason": "Saving dequantized Ministral/Devstral models is different from a FlashAttention failure in Ministral-3." + }, + { + "left": "issue:42490", + "right": "issue:45632", + "accept": false, + "reason": "A warning for local paths ending with a slash is not the same bug as cache-path collisions for leaf-directory names." + }, + { + "left": "issue:43531", + "right": "issue:45600", + "accept": false, + "reason": "Qwen3-MoE sliding_window behavior and Sam3LiteText auto_mappings cleanup are unrelated." + }, + { + "left": "issue:43673", + "right": "issue:44336", + "accept": false, + "reason": "Generation cache missing during chunked prefill is unrelated to ANSI codes in loading_report." + }, + { + "left": "issue:43295", + "right": "issue:45335", + "accept": false, + "reason": "Processor/tokenizer API regression is unrelated to t5gemma resize_token_embeddings not updating decoder.embed_tokens." + }, + { + "left": "issue:44016", + "right": "issue:44062", + "accept": false, + "reason": "Notebook syntax error and AddedToken keyword conflict are different issues." + }, + { + "left": "issue:43526", + "right": "issue:43901", + "accept": false, + "reason": "BeitImageProcessorFast label reduction bug is not the same as stale pipeline docs about return_all_scores." + }, + { + "left": "issue:43824", + "right": "issue:44933", + "accept": false, + "reason": "Missing Qwen2_5_VLForConditionalGeneration import is different from a non-existent image_utils import." + }, + { + "left": "issue:44496", + "right": "issue:45042", + "accept": false, + "reason": "Unrecognized model/config loading and PIL image processors requiring torchvision are unrelated failures." + }, + { + "left": "issue:43335", + "right": "issue:45593", + "accept": false, + "reason": "Sparse layer creation in SwitchTransformersConfig is unrelated to D-FINE auxiliary-loss behavior." + }, + { + "left": "issue:43299", + "right": "issue:45216", + "accept": false, + "reason": "Qwen3VL MoE model loading and Qwen3.5 save_pretrained checkpoint corruption are different bugs." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "Optimizer SGD arguments not being used is unrelated to tie_word_embeddings warnings in Qwen3.5 LoRA fine-tuning." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python 3.9+/3.10+ compatibility is not the same as a general dependency issue." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Gradient checkpointing in eval mode and D-FINE auxiliary-loss behavior are different code paths." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "Documentation drift for text classification is unrelated to unrecognized-model config loading." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "T5Gemma embedding resize bug is unrelated to Whisper empty-transcription generation after align_special_tokens." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "Unrecognized-model/config.json issues are different from AutoModelForCausalLM.from_config failing to unwrap text_config." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Loading failure for Qwen3VL MoE and incorrect saved visual encoder keys for Qwen3.5 are different regressions." + }, + { + "left": "issue:44075", + "right": "issue:44368", + "accept": false, + "reason": "SGD argument handling is unrelated to warning about tie_word_embeddings in Qwen3.5 LoRA." + }, + { + "left": "issue:43976", + "right": "issue:44279", + "accept": false, + "reason": "Python version support and dependency installation problems are not the same bug." + }, + { + "left": "issue:43381", + "right": "issue:45593", + "accept": false, + "reason": "Eval-mode gradient checkpointing failure and D-FINE auxiliary-loss omission do not share a concrete fix." + }, + { + "left": "issue:43901", + "right": "issue:44496", + "accept": false, + "reason": "A docs mismatch in TextClassificationPipeline is unrelated to config/model-type loading errors." + }, + { + "left": "issue:45335", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize behavior in t5gemma does not overlap with Whisper generation on empty transcription." + }, + { + "left": "issue:44496", + "right": "issue:45759", + "accept": false, + "reason": "These are distinct multimodal/config loading issues with different failure points." + }, + { + "left": "issue:43299", + "right": "issue:45357", + "accept": false, + "reason": "Model loading failure and save_pretrained key regression are different underlying issues." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 33, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:33290", + "issue:38175", + "issue:39692", + "issue:41950", + "issue:42994", + "issue:43065", + "issue:43295", + "issue:43299", + "issue:43317", + "issue:43441", + "issue:43450", + "issue:43479", + "issue:43525", + "issue:43577", + "issue:43650", + "issue:43854", + "issue:43901", + "issue:43976", + "issue:44016", + "issue:44242", + "issue:44279", + "issue:44291", + "issue:44297", + "issue:44351", + "issue:44462", + "issue:44466", + "issue:44521", + "issue:44617", + "issue:44683", + "issue:44779", + "issue:44898", + "issue:45020", + "issue:45081", + "issue:45200", + "issue:45216", + "issue:45276", + "issue:45278", + "issue:45325", + "issue:45405", + "issue:45542", + "issue:45600", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:45823", + "issue:45834", + "issue:45907", + "issue:46132" + ], + "soft_pairs": [ + "issue:44242|issue:45200", + "issue:41950|issue:45020", + "issue:44351|issue:45278", + "issue:43479|issue:43525", + "issue:44521|issue:45325", + "issue:44683|issue:45542", + "issue:44779|issue:45812", + "issue:41950|issue:45823", + "issue:43976|issue:45405", + "issue:43577|issue:45216", + "issue:38175|issue:44617", + "issue:44279|issue:45405", + "issue:39692|issue:43901", + "issue:43299|issue:43854", + "issue:43650|issue:46132", + "issue:43295|issue:44779", + "issue:44462|issue:45081", + "issue:45081|issue:45907", + "issue:44297|issue:44462", + "issue:44016|issue:44291", + "issue:41950|issue:43650", + "issue:43450|issue:44279", + "issue:43441|issue:45834", + "issue:45600|issue:45712", + "issue:33290|issue:45656", + "issue:42994|issue:43317", + "issue:44466|issue:45276", + "issue:43065|issue:44898" + ], + "prepared_review_unit_hash": "f54fb0940a09f0cfc644997dcea4e680bedd5d0c94019dc2d148f45d183129d7", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 28, + "serialized_chars": 36032, + "estimated_input_tokens": 9008, + "estimated_eval_tokens": 18272 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "No soft pair looks like a true duplicate; the cluster is mostly a set of unrelated model, tokenizer, config, and infrastructure bugs that only share broad Transformers/Deepspeed similarity.", + "confidence": 0.95, + "canonical_issue_reason": "issue:44521 is the best fallback representative because it is a concrete, actionable bug with strong discussion/inbound linkage, even though it is not a duplicate of the other items.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the clearest overall anchor in this set: active, specific, and more representative than the broader regression reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44242", + "right": "issue:45200", + "accept": false, + "reason": "Different bugs: MoE load-balancing loss vs Gemma4 multimodal config defaults." + }, + { + "left": "issue:41950", + "right": "issue:45020", + "accept": false, + "reason": "Video-classification processor lookup and remote_code breakage are unrelated failures." + }, + { + "left": "issue:44351", + "right": "issue:45278", + "accept": false, + "reason": "One is a missing HybridCache import; the other is a broad import-regression report." + }, + { + "left": "issue:43479", + "right": "issue:43525", + "accept": false, + "reason": "Different config issues: Phi4 multimodal defaults vs missing pad_token_id on Llama4Config." + }, + { + "left": "issue:44521", + "right": "issue:45325", + "accept": false, + "reason": "Multimodal assistant masks and Qwen2.5-VL rope indexing are different code paths." + }, + { + "left": "issue:44683", + "right": "issue:45542", + "accept": false, + "reason": "FlexAttention compilation on torch 2.9 is unrelated to a TensorFlow/tensorboard backend error." + }, + { + "left": "issue:44779", + "right": "issue:45812", + "accept": false, + "reason": "Both mention tokenizer regressions, but they affect different model families and likely different paths." + }, + { + "left": "issue:41950", + "right": "issue:45823", + "accept": false, + "reason": "Video pipeline processor lookup is not the same as Gemma4 device-mismatch during forward." + }, + { + "left": "issue:43976", + "right": "issue:45405", + "accept": false, + "reason": "Python-version compatibility and PEFT version pinning are separate dependency problems." + }, + { + "left": "issue:43577", + "right": "issue:45216", + "accept": false, + "reason": "Blip2 dtype loading and Qwen3.5 save_pretrained serialization are different bugs." + }, + { + "left": "issue:38175", + "right": "issue:44617", + "accept": false, + "reason": "Unexpected zero probabilities in SigLIP2 is unrelated to a Sam3Video OOM." + }, + { + "left": "issue:44279", + "right": "issue:45405", + "accept": false, + "reason": "A generic dependency issue is not the same as an unreleased PEFT minimum-version bump." + }, + { + "left": "issue:39692", + "right": "issue:43901", + "accept": false, + "reason": "Documentation issues, but for different examples, APIs, and failure modes." + }, + { + "left": "issue:43299", + "right": "issue:43854", + "accept": false, + "reason": "Different model-loading failures for different multimodal models." + }, + { + "left": "issue:43650", + "right": "issue:46132", + "accept": false, + "reason": "A placeholder title is unrelated to an AttentionInterface.register behavior change." + }, + { + "left": "issue:43295", + "right": "issue:44779", + "accept": false, + "reason": "Processor/tokenizer regression and Deepseek tokenizer output regression are not the same bug." + }, + { + "left": "issue:44462", + "right": "issue:45081", + "accept": false, + "reason": "Ignoring tokenizer.json and crashing in _patch_mistral_regex are separate tokenizer-loading issues." + }, + { + "left": "issue:45081", + "right": "issue:45907", + "accept": false, + "reason": "A Mistral regex crash is unrelated to list_repo_files failing to forward revision." + }, + { + "left": "issue:44297", + "right": "issue:44462", + "accept": false, + "reason": "Tokenizer class mismatch on save and tokenizer.json being ignored are different save/load defects." + }, + { + "left": "issue:44016", + "right": "issue:44291", + "accept": false, + "reason": "Notebook syntax error and init_empty_weights TypeError are unrelated." + }, + { + "left": "issue:41950", + "right": "issue:43650", + "accept": false, + "reason": "Unrelated video-pipeline bug and placeholder issue title." + }, + { + "left": "issue:43450", + "right": "issue:44279", + "accept": false, + "reason": "Video batch-shape bug is unrelated to a dependency issue." + }, + { + "left": "issue:43441", + "right": "issue:45834", + "accept": false, + "reason": "FlashAttention failure in Ministral and long-ocr index errors in Kosmos2.5 are different model-specific bugs." + }, + { + "left": "issue:45600", + "right": "issue:45712", + "accept": false, + "reason": "Both touch stale code paths, but they concern different files and different CI failures." + }, + { + "left": "issue:33290", + "right": "issue:45656", + "accept": false, + "reason": "Adafactor OOM under DeepSpeed and double optimizer steps under DeepSpeed are not the same defect." + }, + { + "left": "issue:42994", + "right": "issue:43317", + "accept": false, + "reason": "Quantized saving failure and dequantized model loading/offload failure are distinct quantization bugs." + }, + { + "left": "issue:44466", + "right": "issue:45276", + "accept": false, + "reason": "lm_head serialization inconsistency and Gemma4 embedding resizing are different weight-handling issues." + }, + { + "left": "issue:43065", + "right": "issue:44898", + "accept": false, + "reason": "Dummy Conv2d in Sam3PixelDecoder is unrelated to Perceiver resolution/interpolation failure." + } + ] + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:41628", + "issue:42175", + "issue:42491", + "issue:43262", + "issue:43299", + "issue:43441", + "issue:43450", + "issue:43525", + "issue:43643", + "issue:43650", + "issue:43653", + "issue:43673", + "issue:43723", + "issue:43761", + "issue:43784", + "issue:43824", + "issue:43881", + "issue:43901", + "issue:43976", + "issue:43994", + "issue:44016", + "issue:44162", + "issue:44188", + "issue:44242", + "issue:44246", + "issue:44261", + "issue:44263", + "issue:44297", + "issue:44336", + "issue:44361", + "issue:44373", + "issue:44521", + "issue:44589", + "issue:44617", + "issue:44623", + "issue:44671", + "issue:44683", + "issue:44749", + "issue:44779", + "issue:44987", + "issue:45290", + "issue:45362", + "issue:45397", + "issue:45542", + "issue:45561", + "issue:45563", + "issue:45593", + "issue:45823" + ], + "soft_pairs": [ + "issue:43650|issue:45290", + "issue:43653|issue:45593", + "issue:44623|issue:44749", + "issue:43441|issue:44683", + "issue:44261|issue:44263", + "issue:44623|issue:45397", + "issue:43723|issue:44162", + "issue:44016|issue:45823", + "issue:42175|issue:43976", + "issue:44361|issue:45593", + "issue:43262|issue:44521", + "issue:43673|issue:43761", + "issue:44671|issue:44779", + "issue:44987|issue:45542", + "issue:43901|issue:43994", + "issue:43784|issue:45542", + "issue:43824|issue:43976", + "issue:43450|issue:44617", + "issue:43525|issue:43881", + "issue:44297|issue:45397", + "issue:44188|issue:44336", + "issue:43525|issue:43643", + "issue:41628|issue:43976", + "issue:44589|issue:45362", + "issue:43901|issue:44246", + "issue:45561|issue:45563", + "issue:44336|issue:44373", + "issue:42491|issue:43299", + "issue:44242|issue:45593" + ], + "prepared_review_unit_hash": "7b4a5cc65af13c23322aa61f157d0270066e83cd7b0ac989e6098f9802ddb17d", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36072, + "estimated_input_tokens": 9018, + "estimated_eval_tokens": 18292 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is mostly a grab bag of unrelated Transformers bugs; none of the soft pairs are close enough to treat as true duplicates, so I rejected them all. I picked the most substantive/high-signal issue as the canonical representative.", + "confidence": 0.34, + "canonical_issue_reason": "issue:44521 has the clearest, most general bug report with ongoing discussion and multiple references, and it is the strongest representative issue in this mixed set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44521 is the best issue candidate because it is a concrete, user-impacting regression in a core API (`apply_chat_template`) and has the most evidence of active relevance.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43650", + "right": "issue:45290", + "accept": false, + "reason": "Unrelated topics: vague data request vs chat template crash on tool-call assistant messages." + }, + { + "left": "issue:43653", + "right": "issue:45593", + "accept": false, + "reason": "Different models and bugs: tokenizer special-token registration vs missing auxiliary losses in D-FINE." + }, + { + "left": "issue:44623", + "right": "issue:44749", + "accept": false, + "reason": "Different problem class: missing saved processor files vs a performance regression after upgrade." + }, + { + "left": "issue:43441", + "right": "issue:44683", + "accept": false, + "reason": "Both mention attention kernels, but they target different failure modes and code paths (FlashAttention/model loading vs compiled flex_attention on newer torch)." + }, + { + "left": "issue:44261", + "right": "issue:44263", + "accept": false, + "reason": "Different GLM/MoE internals; no clear shared underlying bug." + }, + { + "left": "issue:44623", + "right": "issue:45397", + "accept": false, + "reason": "Processor save file omission vs gemma-4 from_pretrained failure are distinct issues." + }, + { + "left": "issue:43723", + "right": "issue:44162", + "accept": false, + "reason": "Both are loading-related, but one is generic tokenizer loading and the other is an ESM2 model regression; not the same bug." + }, + { + "left": "issue:44016", + "right": "issue:45823", + "accept": false, + "reason": "Notebook syntax error is unrelated to Gemma4 device-mapping failure." + }, + { + "left": "issue:42175", + "right": "issue:43976", + "accept": false, + "reason": "Packaging/backends problem vs Python-version compatibility regression; different root causes." + }, + { + "left": "issue:44361", + "right": "issue:45593", + "accept": false, + "reason": "Tokenizer task AttributeError and D-FINE auxiliary-loss bug are unrelated." + }, + { + "left": "issue:43262", + "right": "issue:44521", + "accept": false, + "reason": "Same API surface (`apply_chat_template`) but different bugs: audio sampling-rate default vs assistant mask generation." + }, + { + "left": "issue:43673", + "right": "issue:43761", + "accept": false, + "reason": "Generation cache issue vs CLIPVision hidden_states regression; different models and code paths." + }, + { + "left": "issue:44671", + "right": "issue:44779", + "accept": false, + "reason": "Different model families and symptoms: CamemBERT MLM predictions vs Deepseek tokenizer output." + }, + { + "left": "issue:44987", + "right": "issue:45542", + "accept": false, + "reason": "Model-loading failure for a specific repo vs missing TensorFlow backend install issue." + }, + { + "left": "issue:43901", + "right": "issue:43994", + "accept": false, + "reason": "Docs mismatch in a pipeline API vs incorrect SigLIP2 inference results." + }, + { + "left": "issue:43784", + "right": "issue:45542", + "accept": false, + "reason": "Import-time NameError in sentence-transformers integration vs TensorFlow backend being absent." + }, + { + "left": "issue:43824", + "right": "issue:43976", + "accept": false, + "reason": "Import error for a specific VL class vs Python-version support regression; not the same bug." + }, + { + "left": "issue:43450", + "right": "issue:44617", + "accept": false, + "reason": "Video processor batching shape bug vs CUDA OOM for Sam3Video." + }, + { + "left": "issue:43525", + "right": "issue:43881", + "accept": false, + "reason": "Different model load failures with no evidence of a shared code-path problem." + }, + { + "left": "issue:44297", + "right": "issue:45397", + "accept": false, + "reason": "Tokenizer save metadata mismatch vs gemma-4 zero3 loading failure are unrelated." + }, + { + "left": "issue:44188", + "right": "issue:44336", + "accept": false, + "reason": "Attention-kernel divergence vs ANSI escape output in loading_report; unrelated." + }, + { + "left": "issue:43525", + "right": "issue:43643", + "accept": false, + "reason": "Both involve missing config fields, but one is a specific model config attr error and the other is a remote-code config loading bug; too different to merge." + }, + { + "left": "issue:41628", + "right": "issue:43976", + "accept": false, + "reason": "Import error for a class export vs Python 3.9/3.10 compatibility issue." + }, + { + "left": "issue:44589", + "right": "issue:45362", + "accept": false, + "reason": "Float8 storage lookup error and Qwen3.5 chat crash are unrelated." + }, + { + "left": "issue:43901", + "right": "issue:44246", + "accept": false, + "reason": "Documentation issue vs intermittent slow import; no shared underlying bug." + }, + { + "left": "issue:45561", + "right": "issue:45563", + "accept": false, + "reason": "xdist race in testing utils vs stale warning in paged generate; different areas." + }, + { + "left": "issue:44336", + "right": "issue:44373", + "accept": false, + "reason": "Loading-report ANSI formatting bug vs wrong docstring; not the same issue." + }, + { + "left": "issue:42491", + "right": "issue:43299", + "accept": false, + "reason": "Both involve Qwen3 MoE loading regressions, but one is a LoRA compatibility problem and the other is a base-model loading breakage; too different for one PR." + }, + { + "left": "issue:44242", + "right": "issue:45593", + "accept": false, + "reason": "Missing load-balancing loss in MoE routing vs D-FINE auxiliary-loss omission; similar wording, different concrete bugs." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 35, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:36246", + "issue:41950", + "issue:42757", + "issue:42886", + "issue:43232", + "issue:43257", + "issue:43262", + "issue:43352", + "issue:43526", + "issue:43531", + "issue:43596", + "issue:43673", + "issue:43901", + "issue:44162", + "issue:44220", + "issue:44242", + "issue:44246", + "issue:44295", + "issue:44336", + "issue:44351", + "issue:44485", + "issue:44496", + "issue:44561", + "issue:44749", + "issue:44779", + "issue:44857", + "issue:44871", + "issue:44877", + "issue:44987", + "issue:44995", + "issue:45003", + "issue:45137", + "issue:45200", + "issue:45250", + "issue:45292", + "issue:45362", + "issue:45375", + "issue:45397", + "issue:45412", + "issue:45459", + "issue:45561", + "issue:45584", + "issue:45701", + "issue:45759", + "issue:45987", + "issue:46082", + "issue:46123", + "issue:46132" + ], + "soft_pairs": [ + "issue:44295|issue:45397", + "issue:44246|issue:44496", + "issue:41950|issue:44220", + "issue:43531|issue:45362", + "issue:42886|issue:46132", + "issue:36246|issue:42757", + "issue:43526|issue:45003", + "issue:36246|issue:44351", + "issue:44336|issue:44485", + "issue:45003|issue:45561", + "issue:43352|issue:45250", + "issue:43673|issue:44561", + "issue:44246|issue:44871", + "issue:44242|issue:44857", + "issue:36246|issue:43596", + "issue:44162|issue:45200", + "issue:44295|issue:44749", + "issue:45412|issue:46123", + "issue:44779|issue:45292", + "issue:43901|issue:45003", + "issue:45292|issue:45584", + "issue:44496|issue:45003", + "issue:45292|issue:45701", + "issue:45987|issue:46082", + "issue:44987|issue:45459", + "issue:44877|issue:45375", + "issue:43232|issue:44336", + "issue:45375|issue:45759", + "issue:43257|issue:45137", + "issue:43262|issue:44246", + "issue:44995|issue:45200" + ], + "prepared_review_unit_hash": "45e97d3b85f02b9481c7a266dac8609ad8188766733eb2e3de503a5f7b738ae4", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 31, + "serialized_chars": 36856, + "estimated_input_tokens": 9214, + "estimated_eval_tokens": 18684 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster does not look like a true duplicate set; the soft pairs mostly connect unrelated model-loading, config, pipeline, cache, and infrastructure issues. I would not merge any of the proposed soft edges.", + "confidence": 0.96, + "canonical_issue_reason": "issue:44246 is the best cluster anchor because it has the broadest symptom description and high discussion activity, making it the most umbrella-like issue among otherwise heterogeneous reports.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44246 is the strongest single issue to represent the set: it is generic, high-activity, and closer to a cross-cutting symptom than the many model-specific failures here.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44295", + "right": "issue:45397", + "accept": false, + "reason": "Different bugs: position_ids buffer access vs. DeepSpeed ZeRO3 from_pretrained loading failure." + }, + { + "left": "issue:44246", + "right": "issue:44496", + "accept": false, + "reason": "Import-time slowdown is unrelated to an unrecognized model/config loading error." + }, + { + "left": "issue:41950", + "right": "issue:44220", + "accept": false, + "reason": "Video-classification processor lookup and audio fbank extraction are different code paths and symptoms." + }, + { + "left": "issue:43531", + "right": "issue:45362", + "accept": false, + "reason": "Both involve Qwen variants, but the reported failures are different concrete issues in different paths." + }, + { + "left": "issue:42886", + "right": "issue:46132", + "accept": false, + "reason": "Offline tokenizer cache loading and AttentionInterface.register behavior are unrelated." + }, + { + "left": "issue:36246", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but for different symbols from different packages and different root causes." + }, + { + "left": "issue:43526", + "right": "issue:45003", + "accept": false, + "reason": "BeitImageProcessorFast label reduction and sys.modules access are unrelated bugs." + }, + { + "left": "issue:36246", + "right": "issue:44351", + "accept": false, + "reason": "Missing Qwen2_5_VLImageProcessor and missing HybridCache are separate import problems." + }, + { + "left": "issue:44336", + "right": "issue:44485", + "accept": false, + "reason": "ANSI output in loading_report is unrelated to GLM-5 RoPE implementation." + }, + { + "left": "issue:45003", + "right": "issue:45561", + "accept": false, + "reason": "Unsafe sys.modules access and pytest-xdist file race are different failure modes." + }, + { + "left": "issue:43352", + "right": "issue:45250", + "accept": false, + "reason": "These only share the Flash Attention 2.0 theme; they do not show the same concrete model/code-path bug." + }, + { + "left": "issue:43673", + "right": "issue:44561", + "accept": false, + "reason": "Chunked-prefill cache handling and trust_remote_code breakage from a removed helper are unrelated." + }, + { + "left": "issue:44246", + "right": "issue:44871", + "accept": false, + "reason": "Import latency and Gemma-3 eos_token_id config mismatch are unrelated." + }, + { + "left": "issue:44242", + "right": "issue:44857", + "accept": false, + "reason": "MoE router loss handling and RT-DETR AMP crash are different model-specific bugs." + }, + { + "left": "issue:36246", + "right": "issue:43596", + "accept": false, + "reason": "Qwen2_5_VLImageProcessor import failure and BertModel ZeRO3 index error are unrelated." + }, + { + "left": "issue:44162", + "right": "issue:45200", + "accept": false, + "reason": "ESM2 breakage and Gemma 4 mm_token_type_ids defaulting are different issues." + }, + { + "left": "issue:44295", + "right": "issue:44749", + "accept": false, + "reason": "position_ids buffer access and slow filtering after upgrade are not the same underlying defect." + }, + { + "left": "issue:45412", + "right": "issue:46123", + "accept": false, + "reason": "RT-DETR memory retention and mask generation partial-batch dropping are unrelated." + }, + { + "left": "issue:44779", + "right": "issue:45292", + "accept": false, + "reason": "Tokenizer correctness regression and resize_token_embeddings output sync are different code paths." + }, + { + "left": "issue:43901", + "right": "issue:45003", + "accept": false, + "reason": "Docs mismatch for return_all_scores and unsafe sys.modules access are unrelated." + }, + { + "left": "issue:45292", + "right": "issue:45584", + "accept": false, + "reason": "Embedding resize propagation and Whisper empty transcription failures do not share a concrete root cause." + }, + { + "left": "issue:44496", + "right": "issue:45003", + "accept": false, + "reason": "Unrecognized model config and sys.modules access are separate loading bugs." + }, + { + "left": "issue:45292", + "right": "issue:45701", + "accept": false, + "reason": "Resize-token-embeddings behavior and version-dependent tokenization changes are not the same bug." + }, + { + "left": "issue:45987", + "right": "issue:46082", + "accept": false, + "reason": "Cache sequence-length return type and LlamaConfig head_dim validation are unrelated." + }, + { + "left": "issue:44987", + "right": "issue:45459", + "accept": false, + "reason": "Model-loading failure on 5.1.0 and protobuf error masking are distinct tokenizer/loading issues." + }, + { + "left": "issue:44877", + "right": "issue:45375", + "accept": false, + "reason": "Strict config rejection for granite_speech and a missing Qwen vision config field are only loosely related by strict config handling, not the same bug." + }, + { + "left": "issue:43232", + "right": "issue:44336", + "accept": false, + "reason": "Generation kwargs sync handling and ANSI codes in loading_report are unrelated." + }, + { + "left": "issue:45375", + "right": "issue:45759", + "accept": false, + "reason": "Missing config field under strict mode and composite Qwen text_config unwrapping are different config-loading bugs." + }, + { + "left": "issue:43257", + "right": "issue:45137", + "accept": false, + "reason": "Qwen3 MOE weight conversion and ZeRO3 deque underflow are different DeepSpeed-related failures." + }, + { + "left": "issue:43262", + "right": "issue:44246", + "accept": false, + "reason": "Audio chat-template sampling-rate default and import-time slowdown are unrelated." + }, + { + "left": "issue:44995", + "right": "issue:45200", + "accept": false, + "reason": "GlmMoeDsa stale cache on second forward and Gemma 4 token-type defaults 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": 36, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:28282", + "issue:30990", + "issue:42886", + "issue:42994", + "issue:43065", + "issue:43232", + "issue:43408", + "issue:43475", + "issue:43482", + "issue:43540", + "issue:43643", + "issue:43646", + "issue:43653", + "issue:43673", + "issue:43931", + "issue:44038", + "issue:44060", + "issue:44077", + "issue:44162", + "issue:44297", + "issue:44361", + "issue:44368", + "issue:44466", + "issue:44623", + "issue:44734", + "issue:44749", + "issue:44869", + "issue:44871", + "issue:44933", + "issue:44977", + "issue:44995", + "issue:45137", + "issue:45229", + "issue:45230", + "issue:45278", + "issue:45310", + "issue:45335", + "issue:45356", + "issue:45381", + "issue:45397", + "issue:45459", + "issue:45563", + "issue:45588", + "issue:45600", + "issue:45656", + "issue:46077", + "issue:46123", + "issue:46132" + ], + "soft_pairs": [ + "issue:43065|issue:43408", + "issue:43475|issue:43540", + "issue:42886|issue:45229", + "issue:43653|issue:44162", + "issue:44995|issue:45588", + "issue:44466|issue:45335", + "issue:44060|issue:44077", + "issue:44368|issue:44977", + "issue:45397|issue:45656", + "issue:44368|issue:45563", + "issue:45137|issue:45600", + "issue:43673|issue:44734", + "issue:43065|issue:46123", + "issue:43931|issue:44368", + "issue:44933|issue:45459", + "issue:44869|issue:45356", + "issue:43232|issue:43673", + "issue:44297|issue:44623", + "issue:44361|issue:44749", + "issue:44871|issue:46132", + "issue:43646|issue:44162", + "issue:43673|issue:45278", + "issue:44623|issue:45656", + "issue:44749|issue:45656", + "issue:44749|issue:45310", + "issue:42994|issue:43482", + "issue:28282|issue:30990", + "issue:43475|issue:45381", + "issue:43475|issue:43643", + "issue:46077|issue:46123", + "issue:44038|issue:44623", + "issue:45230|issue:45397" + ], + "prepared_review_unit_hash": "97110cb9ac456dd0e7df1328222449434f7f7232d996891eeeda9209e0a49bb8", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37133, + "estimated_input_tokens": 9284, + "estimated_eval_tokens": 18824 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These issues are mostly a grab bag of unrelated Transformers regressions across SAM3, Qwen, Gemma, tokenizers, generation, and DeepSpeed. The soft-similarity pairs are thematically related in a few places, but they do not look like the same concrete bug/change, so none should be merged as duplicates.", + "confidence": 0.9, + "canonical_issue_reason": "issue:45278 is the broadest umbrella-style report in the set ('many import errors after update from 4.57.0 to 5.5.0') and is the best single representative if one must be chosen, though the cluster is too heterogeneous to treat as one duplicate family.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most generally useful representative because it describes a wide regression affecting multiple imports after upgrade, which is closer to a cluster-level symptom than the other narrowly scoped bugs.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43065", + "right": "issue:43408", + "accept": false, + "reason": "Both are SAM3-related, but one is about a dummy Conv2d implementation and the other is a model-type/config warning; different bugs." + }, + { + "left": "issue:43475", + "right": "issue:43540", + "accept": false, + "reason": "Different model families and failure modes: missing encoder output field vs video-input processing error." + }, + { + "left": "issue:42886", + "right": "issue:45229", + "accept": false, + "reason": "Offline tokenizer cache loading and multi-GPU CUDA OOM are unrelated." + }, + { + "left": "issue:43653", + "right": "issue:44162", + "accept": false, + "reason": "Tokenizer special-token registration bug vs a broad ESM2 breakage report; not the same code path." + }, + { + "left": "issue:44995", + "right": "issue:45588", + "accept": false, + "reason": "Stale indexer cache crash and flash-attention s_aux=None AttributeError are distinct failures." + }, + { + "left": "issue:44466", + "right": "issue:45335", + "accept": false, + "reason": "Both mention tied/embedding behavior, but one is serialization of lm_head.weight and the other is resize_token_embeddings not updating decoder.embed_tokens." + }, + { + "left": "issue:44060", + "right": "issue:44077", + "accept": false, + "reason": "Both concern model/config warnings, but they describe different models and different underlying problems." + }, + { + "left": "issue:44368", + "right": "issue:44977", + "accept": false, + "reason": "A tie_word_embeddings warning is not the same as a flash-attention generation failure." + }, + { + "left": "issue:45397", + "right": "issue:45656", + "accept": false, + "reason": "Gemma4 from_pretrained/zero3 loading and a double optimizer-step issue in DeepSpeed are separate bugs." + }, + { + "left": "issue:44368", + "right": "issue:45563", + "accept": false, + "reason": "One is a configuration warning about tied embeddings; the other is a stale paged-generate warning." + }, + { + "left": "issue:45137", + "right": "issue:45600", + "accept": false, + "reason": "DeepSpeed ZeRO3 deque/index error vs broken auto_mappings references to removed Sam3LiteText configs; different subsystems and symptoms." + }, + { + "left": "issue:43673", + "right": "issue:44734", + "accept": false, + "reason": "Both involve cache/generation, but one is missing GenerationMixin cache in chunked prefill and the other is a serve-side KV-cache indexing crash." + }, + { + "left": "issue:43065", + "right": "issue:46123", + "accept": false, + "reason": "SAM3 pixel decoder implementation issue vs a MaskGenerationPipeline partial-batch bookkeeping bug; unrelated." + }, + { + "left": "issue:43931", + "right": "issue:44368", + "accept": false, + "reason": "Qwen3-VL-30B weight-shape mismatch is not the same as a tie_word_embeddings warning." + }, + { + "left": "issue:44933", + "right": "issue:45459", + "accept": false, + "reason": "Removed image_utils import vs protobuf-related tokenizer error handling; different modules and failure causes." + }, + { + "left": "issue:44869", + "right": "issue:45356", + "accept": false, + "reason": "Whisper timestamp decode crash and Kimi-K2.5 tokenizer regression are unrelated tokenizer/decoder bugs." + }, + { + "left": "issue:43232", + "right": "issue:43673", + "accept": false, + "reason": "Both touch generation internals, but sync_gpus update_kwargs behavior is not the same as missing cache during chunked prefill." + }, + { + "left": "issue:44297", + "right": "issue:44623", + "accept": false, + "reason": "save_pretrained bugs, but one is tokenizer_class metadata mismatch and the other is processor missing files." + }, + { + "left": "issue:44361", + "right": "issue:44749", + "accept": false, + "reason": "MLukeTokenizer task AttributeError and a performance regression after upgrade are different problems." + }, + { + "left": "issue:44871", + "right": "issue:46132", + "accept": false, + "reason": "Gemma eos_token_id config mismatch is unrelated to AttentionInterface.register changing function behavior." + }, + { + "left": "issue:43646", + "right": "issue:44162", + "accept": false, + "reason": "Generic custom-model initialization breakage and ESM2 breakage are not the same concrete bug." + }, + { + "left": "issue:43673", + "right": "issue:45278", + "accept": false, + "reason": "One is a specific generation cache regression; the other is a broad import-error report after upgrade." + }, + { + "left": "issue:44623", + "right": "issue:45656", + "accept": false, + "reason": "Processor save_pretrained missing files and a deepspeed double-step issue are unrelated." + }, + { + "left": "issue:44749", + "right": "issue:45656", + "accept": false, + "reason": "Performance slowdown after upgrade vs optimizer stepping twice; no shared code-path bug." + }, + { + "left": "issue:44749", + "right": "issue:45310", + "accept": false, + "reason": "Chinese performance regression and Qwen3.5 Moe from_pretrained error are different issues." + }, + { + "left": "issue:42994", + "right": "issue:43482", + "accept": false, + "reason": "Quantized model saving failure vs GGUF loading failure are opposite directions and different code paths." + }, + { + "left": "issue:28282", + "right": "issue:30990", + "accept": false, + "reason": "PyTorch missing import error vs Sentence Transformers getting stuck loading are not the same bug." + }, + { + "left": "issue:43475", + "right": "issue:45381", + "accept": false, + "reason": "Missing fpn_position_embeddings in SAM3 video output and Qwen2.5-VL video position_ids are unrelated video-model bugs." + }, + { + "left": "issue:43475", + "right": "issue:43643", + "accept": false, + "reason": "Different model families and different initialization/config issues." + }, + { + "left": "issue:46077", + "right": "issue:46123", + "accept": false, + "reason": "Gemma4 attention-mask construction and batch-finalization logic in MaskGenerationPipeline are separate bugs." + }, + { + "left": "issue:44038", + "right": "issue:44623", + "accept": false, + "reason": "Qwen3-VL-Moe from_pretrained error and processor.save_pretrained missing files do not describe the same defect." + }, + { + "left": "issue:45230", + "right": "issue:45397", + "accept": false, + "reason": "Generic bug report and Gemma4 zero3 from_pretrained failure are not duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 37, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:30990", + "issue:35141", + "issue:36010", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:42831", + "issue:42898", + "issue:42994", + "issue:43232", + "issue:43377", + "issue:43475", + "issue:43479", + "issue:43504", + "issue:43525", + "issue:43526", + "issue:43673", + "issue:43704", + "issue:43824", + "issue:43873", + "issue:43976", + "issue:44038", + "issue:44079", + "issue:44188", + "issue:44222", + "issue:44242", + "issue:44351", + "issue:44496", + "issue:44568", + "issue:44661", + "issue:44792", + "issue:44869", + "issue:44933", + "issue:44945", + "issue:44964", + "issue:45081", + "issue:45092", + "issue:45103", + "issue:45137", + "issue:45200", + "issue:45310", + "issue:45375", + "issue:45478", + "issue:45561", + "issue:45712", + "issue:45753" + ], + "soft_pairs": [ + "issue:41628|issue:44933", + "issue:42371|issue:45092", + "issue:43475|issue:44661", + "issue:42994|issue:43525", + "issue:45081|issue:45103", + "issue:45375|issue:45478", + "issue:45310|issue:45375", + "issue:43873|issue:44945", + "issue:41628|issue:44351", + "issue:43232|issue:44188", + "issue:42673|issue:43704", + "issue:42831|issue:44222", + "issue:44945|issue:45753", + "issue:36010|issue:41628", + "issue:44222|issue:45310", + "issue:42994|issue:43824", + "issue:42175|issue:45712", + "issue:43673|issue:44188", + "issue:43232|issue:44792", + "issue:44038|issue:45310", + "issue:44964|issue:45137", + "issue:44496|issue:45561", + "issue:43526|issue:45561", + "issue:43504|issue:45561", + "issue:42222|issue:45200", + "issue:43377|issue:44242", + "issue:42175|issue:44933", + "issue:42898|issue:44568", + "issue:43976|issue:44933", + "issue:30990|issue:35141", + "issue:43479|issue:44869", + "issue:43479|issue:44079" + ], + "prepared_review_unit_hash": "81796dffe2fd9d9d6fee8e20ecfd570091b0e810facdf3e4a78124bdeaf886c2", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 37228, + "estimated_input_tokens": 9307, + "estimated_eval_tokens": 18870 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Most pairs are only superficially similar and should stay separate. I tightened the Qwen3.5 triage: 45375/45478 is not a safe duplicate, while the stronger duplicate signal in that cluster is the identical 45310/45478 title pair. Among the provided soft edges, only 42673/43704 is a true duplicate.", + "confidence": 0.84, + "canonical_issue_reason": "issue:45310 is the best issue anchor for the Qwen3.5 MoE from_pretrained regression because it is concrete and its matching duplicate signal is the identical-title report in issue:45478; issue:45375 describes a different missing-field/config serialization problem.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the strongest overall issue candidate because it represents the clearest repeated Qwen3.5 MoE regression and is the cleanest anchor for the true duplicate pair with issue:45478.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41628", + "right": "issue:44933", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols and different code paths." + }, + { + "left": "issue:42371", + "right": "issue:45092", + "accept": false, + "reason": "TF32 configuration behavior and multimodal meta-initialization compatibility are unrelated." + }, + { + "left": "issue:43475", + "right": "issue:44661", + "accept": false, + "reason": "A missing vision-output attribute is not the same bug as a tokenizer-mapping registration failure." + }, + { + "left": "issue:42994", + "right": "issue:43525", + "accept": false, + "reason": "Quantized save failure and missing config attribute are different defects." + }, + { + "left": "issue:45081", + "right": "issue:45103", + "accept": false, + "reason": "Tokenizer regex patching and auto-docstring annotation handling crash in different subsystems." + }, + { + "left": "issue:45375", + "right": "issue:45478", + "accept": false, + "reason": "Related Qwen3.5 reports, but one is a missing deepstack field while the other is a generic from_pretrained error; not the same concrete bug." + }, + { + "left": "issue:45310", + "right": "issue:45375", + "accept": false, + "reason": "Both are Qwen3.5 MoE-related, but the symptoms differ enough that they are not clearly the same bug." + }, + { + "left": "issue:43873", + "right": "issue:44945", + "accept": false, + "reason": "VRAM leakage in dataloader threads and incorrect output under pipeline parallelism are different problems." + }, + { + "left": "issue:41628", + "right": "issue:44351", + "accept": false, + "reason": "They are both imports, but they concern different missing top-level symbols." + }, + { + "left": "issue:43232", + "right": "issue:44188", + "accept": false, + "reason": "Generation kwargs syncing and torch.compile attention-kernel branching are unrelated failures." + }, + { + "left": "issue:42673", + "right": "issue:43704", + "accept": true, + "reason": "Same Qwen3ForCausalLM VRAM leak report with matching title and symptom." + }, + { + "left": "issue:42831", + "right": "issue:44222", + "accept": false, + "reason": "FP8 accuracy degradation and FP8 save_pretrained failure are distinct issues." + }, + { + "left": "issue:44945", + "right": "issue:45753", + "accept": false, + "reason": "Pipeline-parallel output corruption and a Qwen3_5 image-specific infinite loop are different bugs." + }, + { + "left": "issue:36010", + "right": "issue:41628", + "accept": false, + "reason": "GenerationMixin import failure and AutoImageProcessor import failure are separate regressions." + }, + { + "left": "issue:44222", + "right": "issue:45310", + "accept": false, + "reason": "FP8 save_pretrained for moe and Qwen3.5 MoE from_pretrained error are different code paths." + }, + { + "left": "issue:42994", + "right": "issue:43824", + "accept": false, + "reason": "Quantized model saving and Qwen2.5-VL import failure are unrelated." + }, + { + "left": "issue:42175", + "right": "issue:45712", + "accept": false, + "reason": "Missing TF backend from extras and leftover dummy classes leaking into the repo are unrelated packaging issues." + }, + { + "left": "issue:43673", + "right": "issue:44188", + "accept": false, + "reason": "Chunked-prefill cache handling and attention-kernel divergence are not the same bug." + }, + { + "left": "issue:43232", + "right": "issue:44792", + "accept": false, + "reason": "Generation kwargs update and a janus image-generation test failure are different failures." + }, + { + "left": "issue:44038", + "right": "issue:45310", + "accept": false, + "reason": "Both involve Qwen MoE models, but the reported symptoms and model variants differ enough that they are not the same concrete bug." + }, + { + "left": "issue:44964", + "right": "issue:45137", + "accept": false, + "reason": "Phi-4 multimodal loading and a DeepSpeed ZeRO3 deque error are unrelated." + }, + { + "left": "issue:44496", + "right": "issue:45561", + "accept": false, + "reason": "Model config recognition failure and pytest-xdist file race are unrelated." + }, + { + "left": "issue:43526", + "right": "issue:45561", + "accept": false, + "reason": "Label reduction bug and pytest-xdist file race are unrelated." + }, + { + "left": "issue:43504", + "right": "issue:45561", + "accept": false, + "reason": "Legacy-field model loading bug and pytest-xdist file race are unrelated." + }, + { + "left": "issue:42222", + "right": "issue:45200", + "accept": false, + "reason": "Broken VitPose models and Gemma 4 token-type defaults are different model-specific issues." + }, + { + "left": "issue:43377", + "right": "issue:44242", + "accept": false, + "reason": "Padding-mask support in MIMI and missing load-balancing loss when router logits are off are different issues." + }, + { + "left": "issue:42175", + "right": "issue:44933", + "accept": false, + "reason": "Missing TF backend from extras and nonexistent image_utils import are unrelated." + }, + { + "left": "issue:42898", + "right": "issue:44568", + "accept": false, + "reason": "Tokenization cleanup-space behavior and special-token insertion are different regressions." + }, + { + "left": "issue:43976", + "right": "issue:44933", + "accept": false, + "reason": "Python version compatibility and nonexistent image_utils import are unrelated." + }, + { + "left": "issue:30990", + "right": "issue:35141", + "accept": false, + "reason": "A loading stall in Sentence Transformers and token-embedding reinitialization are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44869", + "accept": false, + "reason": "Multimodal config initialization and Whisper word-timestamp decoding are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44079", + "accept": false, + "reason": "Default config initialization and ModelOutput key assignment are unrelated." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The review is not conservative enough on the Qwen3.5 pair. In the packet, issue:45375 is about a missing `deepstack_visual_indexes` field in `Qwen3_5MoeVisionConfig`, while issue:45478 is the generic `Qwen3.5 Moe from_pretrained error`. Those are related, but the evidence does not clearly establish the same concrete bug, so accepting that soft-edge pair overstates the packet. The summary also overstates by saying the only clear duplicate Qwen3.5 pair is represented by 45375/45478; the stronger duplicate signal is the identical 45310/45478 title pair, not 45375/45478." + }, + "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": 38, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:22355", + "issue:28282", + "issue:30333", + "issue:33290", + "issue:34689", + "issue:35141", + "issue:36331", + "issue:41628", + "issue:42757", + "issue:42831", + "issue:42898", + "issue:43116", + "issue:43278", + "issue:43479", + "issue:43644", + "issue:43723", + "issue:43784", + "issue:43824", + "issue:43874", + "issue:44060", + "issue:44188", + "issue:44190", + "issue:44206", + "issue:44222", + "issue:44295", + "issue:44297", + "issue:44373", + "issue:44448", + "issue:44484", + "issue:44568", + "issue:44623", + "issue:44734", + "issue:44779", + "issue:44933", + "issue:45071", + "issue:45092", + "issue:45103", + "issue:45125", + "issue:45137", + "issue:45245", + "issue:45310", + "issue:45356", + "issue:45405", + "issue:45656", + "issue:45712", + "issue:45812", + "issue:46082", + "issue:46132" + ], + "soft_pairs": [ + "issue:44623|issue:45310", + "issue:43278|issue:45137", + "issue:44188|issue:44373", + "issue:42898|issue:45812", + "issue:43784|issue:43824", + "issue:30333|issue:33290", + "issue:44734|issue:45103", + "issue:43723|issue:44568", + "issue:42898|issue:45356", + "issue:42898|issue:44779", + "issue:44060|issue:45125", + "issue:45405|issue:45712", + "issue:43874|issue:45092", + "issue:42898|issue:44448", + "issue:44933|issue:45712", + "issue:44295|issue:45656", + "issue:44448|issue:46132", + "issue:44222|issue:44297", + "issue:43116|issue:44190", + "issue:44297|issue:45656", + "issue:43479|issue:43644", + "issue:44190|issue:46082", + "issue:42831|issue:44623", + "issue:44222|issue:44623", + "issue:45245|issue:46132", + "issue:44206|issue:45103", + "issue:30333|issue:36331", + "issue:41628|issue:42757", + "issue:22355|issue:28282", + "issue:45071|issue:46082", + "issue:30333|issue:34689", + "issue:28282|issue:30333", + "issue:44297|issue:45310", + "issue:30333|issue:35141", + "issue:44222|issue:44484" + ], + "prepared_review_unit_hash": "52da0c376fc795bcf92b5b5b9590089d3e3ae14f38e46322e5af238663f948f4", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 35, + "serialized_chars": 38218, + "estimated_input_tokens": 9555, + "estimated_eval_tokens": 19366 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This set is mostly a loose collection of unrelated Transformers issues, with several soft similarities around tokenizer/v5 regressions but no strong duplicate pairs. The best representative issue is the generic tokenizer regression report, while no PRs are present.", + "confidence": 0.58, + "canonical_issue_reason": "issue:42898 is the broadest, most representative user-facing regression in the set: a concrete v5 behavior change affecting tokenization, which is the main theme shared by several of the softer similarity candidates.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:42898 is the clearest standalone bug report with a specific version boundary and reproducible tokenizer behavior change; it is the most suitable anchor for this cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44623", + "right": "issue:45310", + "accept": false, + "reason": "Both mention save/from_pretrained-style failures, but one is about processor serialization and the other is a Qwen3.5 MoE load error; different code paths and failure modes." + }, + { + "left": "issue:43278", + "right": "issue:45137", + "accept": false, + "reason": "BF16/FP32 dtype drift during train/eval is unrelated to the DeepSpeed ZeRO3 deque IndexError." + }, + { + "left": "issue:44188", + "right": "issue:44373", + "accept": false, + "reason": "Attention-kernel divergence on torch.compile is unrelated to a wrong docstring for position_ids." + }, + { + "left": "issue:42898", + "right": "issue:45812", + "accept": false, + "reason": "Both are tokenizer regressions in v5, but one is a generic clean_up_tokenization_spaces behavior change and the other is Granite token IDs being wrong; different concrete bugs." + }, + { + "left": "issue:43784", + "right": "issue:43824", + "accept": false, + "reason": "Importing sentence-transformers with missing nn is a dependency/import issue, not the missing Qwen2_5_VL model class." + }, + { + "left": "issue:30333", + "right": "issue:33290", + "accept": false, + "reason": "MLflow reporting on failed training jobs is unrelated to an OOM in DeepSpeed with Adafactor." + }, + { + "left": "issue:44734", + "right": "issue:45103", + "accept": false, + "reason": "KV-cache continuation tensor indexing in serve is unrelated to the auto_docstring crash under future annotations." + }, + { + "left": "issue:43723", + "right": "issue:44568", + "accept": false, + "reason": "Both are tokenizer regressions, but one is tokenizer loading in v5 and the other is add_special_tokens/BOS-EOS handling for a specific tokenizer; not the same underlying bug." + }, + { + "left": "issue:42898", + "right": "issue:45356", + "accept": false, + "reason": "Both are tokenizer-related regressions, but one concerns generic tokenization-space cleanup while the other is Kimi-K2.5 codec handling and a warning message." + }, + { + "left": "issue:42898", + "right": "issue:44779", + "accept": false, + "reason": "Deepseek tokenization correctness and clean_up_tokenization_spaces are distinct tokenizer bugs with different symptoms and likely fixes." + }, + { + "left": "issue:44060", + "right": "issue:45125", + "accept": false, + "reason": "A wrong tied-weights warning for Qwen3-Next is unrelated to a missing _tp_plan for Qwen3_5Moe tensor parallelism." + }, + { + "left": "issue:45405", + "right": "issue:45712", + "accept": false, + "reason": "PEFT version pinning and leftover dummy classes are unrelated maintenance issues." + }, + { + "left": "issue:43874", + "right": "issue:45092", + "accept": false, + "reason": "A missing image-patch count method in GLM46V is not the same bug as old InternVL2 checkpoints failing meta initialization." + }, + { + "left": "issue:44933", + "right": "issue:45712", + "accept": false, + "reason": "A nonexistent import from image_utils is unrelated to dummy PT object cleanup." + }, + { + "left": "issue:42898", + "right": "issue:44448", + "accept": false, + "reason": "Pegasus output differences between v4 and v5 are a model-generation regression, not the same as generic tokenization-space cleanup." + }, + { + "left": "issue:44295", + "right": "issue:45656", + "accept": false, + "reason": "Registering position_ids as a buffer causing read errors is unrelated to optimizer.step being called twice with DeepSpeed." + }, + { + "left": "issue:44448", + "right": "issue:46132", + "accept": false, + "reason": "Different output for Pegasus and AttentionInterface.register changing callback behavior are unrelated concerns." + }, + { + "left": "issue:44222", + "right": "issue:44297", + "accept": false, + "reason": "FP8 save_pretrained for MoE models and tokenizer.save_pretrained metadata mismatch are different serialization bugs." + }, + { + "left": "issue:43116", + "right": "issue:44190", + "accept": false, + "reason": "Empty multi-label classification results in an example script is unrelated to loading a local dataset in another training script." + }, + { + "left": "issue:44297", + "right": "issue:45656", + "accept": false, + "reason": "Tokenizer save_pretrained class metadata mismatch is unrelated to DeepSpeed optimizer double-step behavior." + }, + { + "left": "issue:43479", + "right": "issue:43644", + "accept": false, + "reason": "Incorrect default multimodal config initialization is unrelated to non-persistent buffers being filled with junk." + }, + { + "left": "issue:44190", + "right": "issue:46082", + "accept": false, + "reason": "Local dataset loading in an example script is unrelated to LlamaConfig head_dim validation." + }, + { + "left": "issue:42831", + "right": "issue:44623", + "accept": false, + "reason": "FineGrainedFP8 accuracy issues are not the same as missing files from processor.save_pretrained." + }, + { + "left": "issue:44222", + "right": "issue:44623", + "accept": false, + "reason": "Both involve save_pretrained, but one is an FP8 MoE save issue and the other is missing processor files; different artifacts and failure modes." + }, + { + "left": "issue:45245", + "right": "issue:46132", + "accept": false, + "reason": "Category cardinality limits are unrelated to AttentionInterface.register behavior changes." + }, + { + "left": "issue:44206", + "right": "issue:45103", + "accept": false, + "reason": "A LasrFeatureExtractor argument regression is unrelated to an auto_docstring annotation crash." + }, + { + "left": "issue:30333", + "right": "issue:36331", + "accept": false, + "reason": "MLflow reporting for failed jobs and CustomTrainer.compute_loss signature changes are separate training-stack bugs." + }, + { + "left": "issue:41628", + "right": "issue:42757", + "accept": false, + "reason": "Missing AutoImageProcessor import and missing is_offline_mode from huggingface_hub are different import breakages." + }, + { + "left": "issue:22355", + "right": "issue:28282", + "accept": false, + "reason": "Missing transformers.onnx and missing PyTorch for AutoModel are unrelated dependency/import issues." + }, + { + "left": "issue:45071", + "right": "issue:46082", + "accept": false, + "reason": "PretrainedConfig type checking regression is unrelated to LlamaConfig head_dim validation." + }, + { + "left": "issue:30333", + "right": "issue:34689", + "accept": false, + "reason": "MLflow failure reporting and Llama 3.2 vision model loading are unrelated." + }, + { + "left": "issue:28282", + "right": "issue:30333", + "accept": false, + "reason": "PyTorch-not-found ImportError and MLflow reporting on failed training are unrelated." + }, + { + "left": "issue:44297", + "right": "issue:45310", + "accept": false, + "reason": "Tokenizer class metadata mismatch in save_pretrained is unrelated to Qwen3.5 MoE from_pretrained loading failure." + }, + { + "left": "issue:30333", + "right": "issue:35141", + "accept": false, + "reason": "Training-job MLflow reporting and embedding reinitialization after resize are unrelated." + }, + { + "left": "issue:44222", + "right": "issue:44484", + "accept": false, + "reason": "FP8 MoE save_pretrained and the max_shard_size default question are not the same bug." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 39, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:22355", + "issue:28218", + "issue:30990", + "issue:38175", + "issue:39401", + "issue:41628", + "issue:42757", + "issue:42831", + "issue:42886", + "issue:42907", + "issue:42994", + "issue:43066", + "issue:43232", + "issue:43479", + "issue:43723", + "issue:43756", + "issue:43761", + "issue:43824", + "issue:44038", + "issue:44060", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44360", + "issue:44373", + "issue:44442", + "issue:44483", + "issue:44485", + "issue:44514", + "issue:44568", + "issue:44623", + "issue:44829", + "issue:44871", + "issue:44913", + "issue:44977", + "issue:44991", + "issue:45092", + "issue:45103", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45406", + "issue:45593", + "issue:45753", + "issue:45834", + "issue:46002", + "issue:46050", + "issue:46082" + ], + "soft_pairs": [ + "issue:43723|issue:44977", + "issue:43232|issue:45834", + "issue:44263|issue:45092", + "issue:43761|issue:46082", + "issue:44373|issue:44485", + "issue:44483|issue:45406", + "issue:44263|issue:44360", + "issue:44295|issue:44623", + "issue:44514|issue:45103", + "issue:43723|issue:44871", + "issue:45092|issue:45357", + "issue:43723|issue:44991", + "issue:30990|issue:38175", + "issue:42907|issue:46050", + "issue:44829|issue:46002", + "issue:43723|issue:44442", + "issue:44261|issue:45092", + "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:42831|issue:45753", + "issue:43756|issue:44060", + "issue:42886|issue:43066", + "issue:22355|issue:28218" + ], + "prepared_review_unit_hash": "ce4ec132738576a0cc51da25d7412123a518cc12d61c46434caedad4bdcbc33a", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 30, + "serialized_chars": 36333, + "estimated_input_tokens": 9084, + "estimated_eval_tokens": 18424 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "Mostly unrelated issue reports with a loose concentration of tokenizer/import regressions; none of the soft pairs are strong enough to treat as true duplicates from titles alone.", + "confidence": 0.42, + "canonical_issue_reason": "issue:43723 is the broadest and most representative of the repeated v5 tokenizer-loading regression theme, so it is the best single issue anchor in this mixed set.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:43723 best fits the recurring tokenizer-loading/import failure pattern, but the cluster is too heterogeneous for a strong canonical choice.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43723", + "right": "issue:44977", + "accept": false, + "reason": "One is a tokenizer-loading regression; the other is a Qwen3.5 generation/flash-attention problem. Different code paths and symptoms." + }, + { + "left": "issue:43232", + "right": "issue:45834", + "accept": false, + "reason": "Generation kwargs handling after sync_gpus is unrelated to Kosmos2.5 long OCR input indexing." + }, + { + "left": "issue:44263", + "right": "issue:45092", + "accept": false, + "reason": "torch.split/indexer behavior and InternVL2 meta-init compatibility are unrelated bugs." + }, + { + "left": "issue:43761", + "right": "issue:46082", + "accept": false, + "reason": "CLIPVisionModel hidden_states regression and LlamaConfig head_dim validation are different subsystems and failures." + }, + { + "left": "issue:44373", + "right": "issue:44485", + "accept": false, + "reason": "A docstring error for position_ids is not the same as a GLM-5 RoPE implementation bug." + }, + { + "left": "issue:44483", + "right": "issue:45406", + "accept": false, + "reason": "Chat/completions request validation and Gemma4Processor missing _tokenizer are distinct problems." + }, + { + "left": "issue:44263", + "right": "issue:44360", + "accept": false, + "reason": "Both touch the DSA indexer, but one is torch.split return values while the other is a missing ReLU; not the same bug." + }, + { + "left": "issue:44295", + "right": "issue:44623", + "accept": false, + "reason": "Reading position_ids after buffer registration and processor.save_pretrained missing files are different failure modes." + }, + { + "left": "issue:44514", + "right": "issue:45103", + "accept": false, + "reason": "Qwen2_5_VLProcessor batch padding crash is unrelated to the auto_docstring future-annotations AttributeError." + }, + { + "left": "issue:43723", + "right": "issue:44871", + "accept": false, + "reason": "Tokenizer-loading regression vs mismatched eos_token_id configuration; related to tokenizers, but not the same underlying bug." + }, + { + "left": "issue:45092", + "right": "issue:45357", + "accept": false, + "reason": "InternVL2 meta-init incompatibility and Qwen3.5 save_pretrained key corruption are separate model-specific save/load issues." + }, + { + "left": "issue:43723", + "right": "issue:44991", + "accept": false, + "reason": "Both are tokenizer-load failures in v5, but one is generic and the other is model-specific; insufficient evidence of one shared root cause." + }, + { + "left": "issue:30990", + "right": "issue:38175", + "accept": false, + "reason": "Sentence Transformers loading hang and SigLIP2 zero-probabilities are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:46050", + "accept": false, + "reason": "Different quantization save/load problems with different methods and symptoms; not clearly the same bug." + }, + { + "left": "issue:44829", + "right": "issue:46002", + "accept": false, + "reason": "Flash-attention training degeneration and NaNs after checkpoint/init are not the same code-path issue." + }, + { + "left": "issue:43723", + "right": "issue:44442", + "accept": false, + "reason": "Both involve AutoTokenizer loading, but the metadata is too different to conclude a single duplicate bug." + }, + { + "left": "issue:44261", + "right": "issue:45092", + "accept": false, + "reason": "MLA layernorm epsilon precision and InternVL2 meta-init are unrelated." + }, + { + "left": "issue:41628", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but for different symbols and likely different root causes; not a true duplicate." + }, + { + "left": "issue:42831", + "right": "issue:45310", + "accept": false, + "reason": "FineGrainedFP8 accuracy and Qwen3.5 from_pretrained error are unrelated." + }, + { + "left": "issue:43479", + "right": "issue:44913", + "accept": false, + "reason": "Config default initialization on None and rotary_pct not persisting on reload are different config bugs." + }, + { + "left": "issue:44263", + "right": "issue:44373", + "accept": false, + "reason": "Indexing/torch.split behavior and a docstring typo are not the same underlying issue." + }, + { + "left": "issue:42831", + "right": "issue:44038", + "accept": false, + "reason": "Accuracy with FineGrainedFP8 and a Qwen3-VL-Moe bug are not the same change or failure." + }, + { + "left": "issue:39401", + "right": "issue:44568", + "accept": false, + "reason": "Offset mapping wrong for Qwen3 and add_special_tokens missing BOS/EOS are distinct tokenizer bugs." + }, + { + "left": "issue:42757", + "right": "issue:45278", + "accept": false, + "reason": "A missing huggingface_hub symbol and broad import errors after upgrade are related only at a very high level; not enough for duplication." + }, + { + "left": "issue:43232", + "right": "issue:45593", + "accept": false, + "reason": "Generation-state handling 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 workflows and bugs." + }, + { + "left": "issue:42831", + "right": "issue:45753", + "accept": false, + "reason": "FineGrainedFP8 accuracy and Qwen3_5 infinite-loop image behavior are unrelated." + }, + { + "left": "issue:43756", + "right": "issue:44060", + "accept": false, + "reason": "Smollm3 RoPE-layer dropping and Qwen3-Next tied-weights warning are different model/config issues." + }, + { + "left": "issue:42886", + "right": "issue:43066", + "accept": false, + "reason": "Offline cache loading failure and wrong decoder type in v5 are different tokenizer issues." + }, + { + "left": "issue:22355", + "right": "issue:28218", + "accept": false, + "reason": "Missing transformers.onnx module and extra-space tokenization behavior are unrelated." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 40, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:29127", + "issue:30990", + "issue:33357", + "issue:36296", + "issue:39401", + "issue:41628", + "issue:42371", + "issue:42907", + "issue:43116", + "issue:43122", + "issue:43232", + "issue:43408", + "issue:43452", + "issue:43475", + "issue:43576", + "issue:43646", + "issue:43784", + "issue:43824", + "issue:43937", + "issue:43976", + "issue:44038", + "issue:44117", + "issue:44222", + "issue:44261", + "issue:44263", + "issue:44295", + "issue:44297", + "issue:44351", + "issue:44360", + "issue:44373", + "issue:44485", + "issue:44488", + "issue:44514", + "issue:44521", + "issue:44625", + "issue:44671", + "issue:44704", + "issue:44743", + "issue:44779", + "issue:44829", + "issue:44945", + "issue:44964", + "issue:45200", + "issue:45278", + "issue:45290", + "issue:45600", + "issue:46002", + "issue:46132" + ], + "soft_pairs": [ + "issue:43122|issue:44521", + "issue:43576|issue:43784", + "issue:44671|issue:46132", + "issue:44117|issue:44704", + "issue:43646|issue:46002", + "issue:30990|issue:33357", + "issue:42371|issue:44263", + "issue:44704|issue:45290", + "issue:43232|issue:44743", + "issue:41628|issue:45278", + "issue:44295|issue:44297", + "issue:42907|issue:44945", + "issue:43122|issue:44514", + "issue:43976|issue:44945", + "issue:39401|issue:44488", + "issue:44261|issue:44373", + "issue:43784|issue:45600", + "issue:42907|issue:43937", + "issue:43408|issue:43475", + "issue:44360|issue:44485", + "issue:39401|issue:44521", + "issue:44625|issue:44829", + "issue:43116|issue:45200", + "issue:42907|issue:44964", + "issue:44038|issue:44297", + "issue:29127|issue:30990", + "issue:43122|issue:44779", + "issue:36296|issue:43116", + "issue:43824|issue:44351", + "issue:44038|issue:44295", + "issue:42907|issue:43452", + "issue:44038|issue:44222" + ], + "prepared_review_unit_hash": "e228d15e2d0065acabe271c1bbaa4f26f0acb70dc14c6a261c4ad50b7f1fb170", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 32, + "serialized_chars": 36845, + "estimated_input_tokens": 9212, + "estimated_eval_tokens": 18680 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: most pairs are only loosely related by broad Transformers subsystems or version-regression language, not the same underlying bug/change. No soft-edge pair looks safe to merge as a duplicate.", + "confidence": 0.9, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the broadest regression-style report in the set (many import errors after upgrading to v5.5.0) and is the best loose representative if a single issue must anchor the cluster, but it does not match most of the other items closely.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43122", + "right": "issue:44521", + "accept": false, + "reason": "Both involve tokenization/template behavior, but one is a version-to-version tokenization change and the other is multimodal assistant-mask generation; different bugs." + }, + { + "left": "issue:43576", + "right": "issue:43784", + "accept": false, + "reason": "A broken `transformers env` command and an import-time `nn` NameError are unrelated failure modes." + }, + { + "left": "issue:44671", + "right": "issue:46132", + "accept": false, + "reason": "Incorrect CamemBERT MLM predictions and `AttentionInterface.register` changing function behavior are distinct code paths." + }, + { + "left": "issue:44117", + "right": "issue:44704", + "accept": false, + "reason": "Tokenizer mapping fallback handling vs `AutoProcessor.from_pretrained` kwargs forwarding are separate API issues." + }, + { + "left": "issue:43646", + "right": "issue:46002", + "accept": false, + "reason": "Custom model initialization breakage and NaNs in classification heads are different underlying problems." + }, + { + "left": "issue:30990", + "right": "issue:33357", + "accept": false, + "reason": "Loading hangs for Sentence Transformers and a MacOS bus error with a CLIP model are not the same bug." + }, + { + "left": "issue:42371", + "right": "issue:44263", + "accept": false, + "reason": "TF32 API guidance and a `torch.split()` return-value issue in GlmMoeDsaIndexer are unrelated." + }, + { + "left": "issue:44704", + "right": "issue:45290", + "accept": false, + "reason": "`cached_file` kwarg forwarding and chat-template crashes with tool-call messages are different processor paths." + }, + { + "left": "issue:43232", + "right": "issue:44743", + "accept": false, + "reason": "Both touch generation/cache logic, but one is `_update_model_kwargs_for_generation` after `sync_gpus` and the other is recurrent-state reset in a Qwen module; not the same defect." + }, + { + "left": "issue:41628", + "right": "issue:45278", + "accept": false, + "reason": "A single missing `AutoImageProcessor` import is narrower than the broad multi-import regression in v5.5.0." + }, + { + "left": "issue:44295", + "right": "issue:44297", + "accept": false, + "reason": "Reading `position_ids` after buffer registration and saving tokenizer class metadata are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:44945", + "accept": false, + "reason": "Dequantized model saving and incorrect output under pipeline parallelism are different functional areas." + }, + { + "left": "issue:43122", + "right": "issue:44514", + "accept": false, + "reason": "Both mention tokenization/chat templates, but the first is a general version tokenization shift and the second is a batched multimodal padding crash." + }, + { + "left": "issue:43976", + "right": "issue:44945", + "accept": false, + "reason": "Python version compatibility and pipeline-parallel inference correctness are unrelated." + }, + { + "left": "issue:39401", + "right": "issue:44488", + "accept": false, + "reason": "Qwen3 offset mapping errors and a model load failure for `cjvt/sleng-bert` do not describe the same bug." + }, + { + "left": "issue:44261", + "right": "issue:44373", + "accept": false, + "reason": "A numerical precision/config issue and a wrong docstring are not duplicates." + }, + { + "left": "issue:43784", + "right": "issue:45600", + "accept": false, + "reason": "Import-time `nn` failure in sentence-transformers and stale `auto_mappings.py` config references are different import regressions." + }, + { + "left": "issue:42907", + "right": "issue:43937", + "accept": false, + "reason": "Saving dequantized models and invalid `GenerationConfig` errors are separate problems." + }, + { + "left": "issue:43408", + "right": "issue:43475", + "accept": false, + "reason": "The first is a model-type mismatch warning; the second is a missing attribute on a SAM3 output object." + }, + { + "left": "issue:44360", + "right": "issue:44485", + "accept": false, + "reason": "A missing ReLU in the DSA indexer and GLM-5 RoPE implementation issues are unrelated." + }, + { + "left": "issue:39401", + "right": "issue:44521", + "accept": false, + "reason": "Offset-mapping correctness and multimodal assistant masks are different tokenizer/processor bugs." + }, + { + "left": "issue:44625", + "right": "issue:44829", + "accept": false, + "reason": "`num_labels` propagation for Qwen3.5 and flash-attention-3 degenerate training are separate model/training issues." + }, + { + "left": "issue:43116", + "right": "issue:45200", + "accept": false, + "reason": "Empty multi-label outputs in an example script and missing default `mm_token_type_ids` are not the same failure." + }, + { + "left": "issue:42907", + "right": "issue:44964", + "accept": false, + "reason": "Dequantized save failures and inability to load a multimodal Phi-4 model are different code paths." + }, + { + "left": "issue:44038", + "right": "issue:44297", + "accept": false, + "reason": "Qwen3-VL-Moe behavior and tokenizer save metadata mismatch are unrelated." + }, + { + "left": "issue:29127", + "right": "issue:30990", + "accept": false, + "reason": "LayoutLMv3 box-validation messaging and a loading hang in Sentence Transformers are unrelated." + }, + { + "left": "issue:43122", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they affect different models and different symptoms (`offset_mapping` vs general incorrect tokenization)." + }, + { + "left": "issue:36296", + "right": "issue:43116", + "accept": false, + "reason": "Tensor-parallel training bug and example-script multi-label output failure are different areas." + }, + { + "left": "issue:43824", + "right": "issue:44351", + "accept": false, + "reason": "Missing Qwen2.5-VL export and missing HybridCache export are separate import errors." + }, + { + "left": "issue:44038", + "right": "issue:44295", + "accept": false, + "reason": "Qwen3-VL-Moe model behavior and `position_ids` buffer-read issues are unrelated." + }, + { + "left": "issue:42907", + "right": "issue:43452", + "accept": false, + "reason": "Saving dequantized models and `gguf_file`-related loading failures are different bugs." + }, + { + "left": "issue:44038", + "right": "issue:44222", + "accept": false, + "reason": "A Qwen3-VL-Moe regression and FP8 save_pretrained behavior for MoE are not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 41, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:22355", + "issue:29127", + "issue:29159", + "issue:33453", + "issue:38617", + "issue:39401", + "issue:42222", + "issue:42371", + "issue:42673", + "issue:42757", + "issue:43064", + "issue:43066", + "issue:43116", + "issue:43122", + "issue:43278", + "issue:43441", + "issue:43493", + "issue:43611", + "issue:43673", + "issue:43784", + "issue:43825", + "issue:43827", + "issue:43874", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44075", + "issue:44351", + "issue:44380", + "issue:44485", + "issue:44704", + "issue:44779", + "issue:44936", + "issue:44945", + "issue:44977", + "issue:45072", + "issue:45083", + "issue:45216", + "issue:45278", + "issue:45357", + "issue:45446", + "issue:45478", + "issue:45563", + "issue:45753", + "issue:46002", + "issue:46077" + ], + "soft_pairs": [ + "issue:45216|issue:45478", + "issue:42757|issue:44351", + "issue:43278|issue:44945", + "issue:43784|issue:45446", + "issue:43122|issue:44977", + "issue:22355|issue:29159", + "issue:43493|issue:45753", + "issue:43066|issue:44779", + "issue:43992|issue:44704", + "issue:38617|issue:45278", + "issue:42673|issue:45072", + "issue:43937|issue:44485", + "issue:22355|issue:29127", + "issue:43611|issue:46002", + "issue:43976|issue:45446", + "issue:43116|issue:43493", + "issue:42222|issue:43441", + "issue:44075|issue:44936", + "issue:42371|issue:43874", + "issue:43064|issue:43673", + "issue:45357|issue:45478", + "issue:44038|issue:45478", + "issue:43116|issue:43825", + "issue:45083|issue:45446", + "issue:43937|issue:45563", + "issue:44380|issue:46077", + "issue:33453|issue:39401", + "issue:43827|issue:45446" + ], + "prepared_review_unit_hash": "ab96bd57220bbb15f12737054938a1d10b9f9d833384787b1d3814418ff382a4", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 28, + "serialized_chars": 35097, + "estimated_input_tokens": 8775, + "estimated_eval_tokens": 17806 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is mostly heterogeneous: a few tokenizer/import/regression themes recur, but the soft pairs are not the same concrete bug/change. No PRs are present. The best issue representative is the broad post-upgrade import-error report.", + "confidence": 0.93, + "canonical_issue_reason": "issue:45278 is the closest umbrella report among the set: it captures a broad post-upgrade import-error regression, making it the most representative issue in an otherwise mixed cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the most broadly representative and can plausibly subsume multiple import-error reports, even though it is not a true duplicate of the specific symbol-level failures.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45216", + "right": "issue:45478", + "accept": false, + "reason": "Same model family, but one is a save_pretrained checkpoint-key regression and the other is a from_pretrained load error; related area, not the same concrete bug." + }, + { + "left": "issue:42757", + "right": "issue:44351", + "accept": false, + "reason": "Both are import errors, but for different symbols from different packages/modules; no shared code-path bug is evident." + }, + { + "left": "issue:43278", + "right": "issue:44945", + "accept": false, + "reason": "Both touch training/eval behavior, but one is a dtype shift and the other is incorrect output under pipeline parallelism; too different to duplicate." + }, + { + "left": "issue:43784", + "right": "issue:45446", + "accept": false, + "reason": "Unrelated import/version problems: sentence-transformers import NameError versus a flex_attention version check bug." + }, + { + "left": "issue:43122", + "right": "issue:44977", + "accept": false, + "reason": "Both are v5 regressions involving tokenization/generation, but the concrete failures differ: tokenizer output drift vs flash-attention generation failure." + }, + { + "left": "issue:22355", + "right": "issue:29159", + "accept": false, + "reason": "Module import failure for transformers.onnx is unrelated to slow/fast tokenizer inconsistency." + }, + { + "left": "issue:43493", + "right": "issue:45753", + "accept": false, + "reason": "Different models and failure modes: SigLIP2 implementation discrepancy versus an infinite loop on a specific image." + }, + { + "left": "issue:43066", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer-related regressions, but one is about decoder type selection in v5 and the other is Deepseek output correctness; not the same bug." + }, + { + "left": "issue:43992", + "right": "issue:44704", + "accept": false, + "reason": "Two separate from_pretrained-related bugs in different classes; one misses a weight key, the other drops kwargs to cached_file." + }, + { + "left": "issue:38617", + "right": "issue:45278", + "accept": false, + "reason": "45278 is a broad import-error umbrella, but 38617 is a specific missing symbol in configuration_utils; not clearly the same underlying defect." + }, + { + "left": "issue:42673", + "right": "issue:45072", + "accept": false, + "reason": "VRAM leak in Qwen3ForCausalLM threads and bfloat16 dtype mismatches are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:44485", + "accept": false, + "reason": "Both mention GLM-5, but one is invalid GenerationConfig and the other is a RoPE implementation discussion; different issues." + }, + { + "left": "issue:22355", + "right": "issue:29127", + "accept": false, + "reason": "Missing transformers.onnx module and a LayoutLMv3 error-message clarity issue are not the same bug." + }, + { + "left": "issue:43611", + "right": "issue:46002", + "accept": false, + "reason": "Model loading via base_model_prefix and NaNs in classification heads are separate failure modes." + }, + { + "left": "issue:43976", + "right": "issue:45446", + "accept": false, + "reason": "Python version support breakage and a PyTorch version-check import issue are unrelated." + }, + { + "left": "issue:43116", + "right": "issue:43493", + "accept": false, + "reason": "Different subsystems entirely: run_classification multi-label output vs SigLIP2 model discrepancy." + }, + { + "left": "issue:42222", + "right": "issue:43441", + "accept": false, + "reason": "VitPose import breakage and Ministral-3 FlashAttention failure are unrelated model-specific bugs." + }, + { + "left": "issue:44075", + "right": "issue:44936", + "accept": false, + "reason": "Optimizer SGD arguments being ignored is not the same as trainer.evaluate failing after train." + }, + { + "left": "issue:42371", + "right": "issue:43874", + "accept": false, + "reason": "TF32 API guidance and a missing image-processor helper method are unrelated." + }, + { + "left": "issue:43064", + "right": "issue:43673", + "accept": false, + "reason": "Optimizer-state corruption under FSDP2/PEFT is unrelated to generation cache handling in chunked_prefill." + }, + { + "left": "issue:45357", + "right": "issue:45478", + "accept": false, + "reason": "Both concern Qwen3.5 save/load behavior, but one is about incorrect saved visual encoder keys and the other is a from_pretrained error; related, but not clearly the same concrete bug." + }, + { + "left": "issue:44038", + "right": "issue:45478", + "accept": false, + "reason": "Same broad Qwen3 family, but different model variants and failure descriptions; not enough to treat as one bug." + }, + { + "left": "issue:43116", + "right": "issue:43825", + "accept": false, + "reason": "A classification script bug and a pipeline() translation-task error message issue are unrelated." + }, + { + "left": "issue:45083", + "right": "issue:45446", + "accept": false, + "reason": "Qwen3 Omni-MoE feature-length helper behavior and a flex_attention import/version check are unrelated." + }, + { + "left": "issue:43937", + "right": "issue:45563", + "accept": false, + "reason": "GenerationConfig validation failure and a stale num_return_sequences warning are different generate-path issues." + }, + { + "left": "issue:44380", + "right": "issue:46077", + "accept": false, + "reason": "Both concern attention behavior, but GPT-2 backend scaling and Gemma4 bidirectional masking are distinct code paths and fixes." + }, + { + "left": "issue:33453", + "right": "issue:39401", + "accept": false, + "reason": "Tokenizer loading regression and Qwen3 offset_mapping correctness are related only at a high level; not the same bug." + }, + { + "left": "issue:43827", + "right": "issue:45446", + "accept": false, + "reason": "Docs still referencing pipeline() and a PyTorch version-check import problem 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": 42, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:38617", + "issue:39401", + "issue:41628", + "issue:42175", + "issue:42222", + "issue:42371", + "issue:42617", + "issue:42673", + "issue:42831", + "issue:43066", + "issue:43122", + "issue:43408", + "issue:43421", + "issue:43450", + "issue:43577", + "issue:43582", + "issue:43742", + "issue:43906", + "issue:43937", + "issue:43986", + "issue:44075", + "issue:44117", + "issue:44220", + "issue:44261", + "issue:44291", + "issue:44410", + "issue:44466", + "issue:44483", + "issue:44484", + "issue:44496", + "issue:44779", + "issue:44829", + "issue:44933", + "issue:44977", + "issue:45071", + "issue:45081", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45341", + "issue:45405", + "issue:45446", + "issue:45464", + "issue:45542", + "issue:45561", + "issue:45588", + "issue:45656", + "issue:46002" + ], + "soft_pairs": [ + "issue:44117|issue:45588", + "issue:39401|issue:44779", + "issue:42673|issue:43906", + "issue:43937|issue:44977", + "issue:44466|issue:46002", + "issue:44291|issue:45071", + "issue:43450|issue:43986", + "issue:45405|issue:45446", + "issue:45341|issue:45561", + "issue:44117|issue:45081", + "issue:38617|issue:41628", + "issue:44410|issue:45310", + "issue:44829|issue:45216", + "issue:39401|issue:43421", + "issue:42673|issue:43582", + "issue:44484|issue:45310", + "issue:42617|issue:45341", + "issue:44933|issue:45278", + "issue:42371|issue:44261", + "issue:42673|issue:43577", + "issue:42222|issue:45542", + "issue:43408|issue:43742", + "issue:38617|issue:44220", + "issue:43937|issue:44496", + "issue:42175|issue:42222", + "issue:44483|issue:45464", + "issue:44075|issue:45656", + "issue:43066|issue:43122", + "issue:42371|issue:42831" + ], + "prepared_review_unit_hash": "223ed6cc879294fb89be4664e7ae783b5d66bc53be6b2b0f669015c45acc081b", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 29, + "serialized_chars": 36230, + "estimated_input_tokens": 9058, + "estimated_eval_tokens": 18372 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is very heterogeneous: most soft matches are superficial similarities across unrelated Transformers issues (tokenizers, model loading, imports, optimizer behavior, testing, and API regressions). I did not find any pair that clearly looks like the same underlying bug or change.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the closest single representative because it captures a broad import-regression theme, but the set is too mixed to treat it as a true canonical duplicate hub.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44117", + "right": "issue:45588", + "accept": false, + "reason": "TOKENIZER_MAPPING_NAMES returning None and a flash_attention s_aux=None crash are different code paths and failure modes." + }, + { + "left": "issue:39401", + "right": "issue:44779", + "accept": false, + "reason": "Both are tokenizer regressions, but they concern different models and different symptoms (offset mapping vs incorrect generation/tokenization behavior)." + }, + { + "left": "issue:42673", + "right": "issue:43906", + "accept": false, + "reason": "VRAM leak in multi-threaded dataloaders is unrelated to an isolated reproduction of another issue." + }, + { + "left": "issue:43937", + "right": "issue:44977", + "accept": false, + "reason": "GenerationConfig invalid and flash-attention generation failure are distinct bugs with different apparent causes." + }, + { + "left": "issue:44466", + "right": "issue:46002", + "accept": false, + "reason": "lm_head.weight serialization inconsistency is not the same as NaNs in classification heads during checkout/init." + }, + { + "left": "issue:44291", + "right": "issue:45071", + "accept": false, + "reason": "Unexpected _is_hf_initialized argument and PretrainedConfig type checking are separate regression points." + }, + { + "left": "issue:43450", + "right": "issue:43986", + "accept": false, + "reason": "Batched video processor shape bug and missing torchvision AutoProcessor crash are different issues." + }, + { + "left": "issue:45405", + "right": "issue:45446", + "accept": false, + "reason": "A pinned PEFT version not yet released and a PyTorch version check bug are unrelated dependency problems." + }, + { + "left": "issue:45341", + "right": "issue:45561", + "accept": false, + "reason": "A generic testing_utils bug is not the same as the captured_info.txt race under pytest-xdist." + }, + { + "left": "issue:44117", + "right": "issue:45081", + "accept": false, + "reason": "Tokenizer mapping None and Mistral regex patch backend_tokenizer AttributeError are different tokenizer failures." + }, + { + "left": "issue:38617", + "right": "issue:41628", + "accept": false, + "reason": "Different missing imports in different APIs; no shared underlying defect is evident." + }, + { + "left": "issue:44410", + "right": "issue:45310", + "accept": false, + "reason": "Missing attn projections in qwen3next and Qwen3.5 MoE from_pretrained failure are different model-specific bugs." + }, + { + "left": "issue:44829", + "right": "issue:45216", + "accept": false, + "reason": "Flash-attention 3 training degeneration and save_pretrained checkpoint corruption are not the same concrete bug." + }, + { + "left": "issue:39401", + "right": "issue:43421", + "accept": false, + "reason": "Offset mapping wrong and post-processor not updating on runtime special-token changes are related to tokenizers but not the same problem." + }, + { + "left": "issue:42673", + "right": "issue:43582", + "accept": false, + "reason": "A VRAM leak in dataloader threads is unrelated to an Apple Silicon caching_allocator_warmup TypeError." + }, + { + "left": "issue:44484", + "right": "issue:45310", + "accept": false, + "reason": "A max_shard_size question is unrelated to a Qwen3.5 MoE from_pretrained error." + }, + { + "left": "issue:42617", + "right": "issue:45341", + "accept": false, + "reason": "3d_parallel.py execution failure and a testing_utils bug are different issues." + }, + { + "left": "issue:44933", + "right": "issue:45278", + "accept": false, + "reason": "Both mention import problems, but one is a specific missing import and the other is a broad multi-import regression; not enough evidence they are the same bug." + }, + { + "left": "issue:42371", + "right": "issue:44261", + "accept": false, + "reason": "TF32 API guidance and MLA rms_norm_eps precision error are unrelated." + }, + { + "left": "issue:42673", + "right": "issue:43577", + "accept": false, + "reason": "VRAM leakage and Blip2 dtype remaining float32 are different bugs." + }, + { + "left": "issue:42222", + "right": "issue:45542", + "accept": false, + "reason": "Broken vitpose models and a tensorboard/TensorFlow backend error are unrelated." + }, + { + "left": "issue:43408", + "right": "issue:43742", + "accept": false, + "reason": "SAM3 model-type warning and MobileLLM key error are distinct loading issues." + }, + { + "left": "issue:38617", + "right": "issue:44220", + "accept": false, + "reason": "layer_type_validation import failure and _torch_extract_fbank_features() issue are not the same code-path problem." + }, + { + "left": "issue:43937", + "right": "issue:44496", + "accept": false, + "reason": "Invalid GenerationConfig and unrecognized model/model_type missing are different failure classes." + }, + { + "left": "issue:42175", + "right": "issue:42222", + "accept": false, + "reason": "TensorFlow backend packaging and vitpose model breakage are unrelated." + }, + { + "left": "issue:44483", + "right": "issue:45464", + "accept": false, + "reason": "A /v1/chat/completions API incompatibility and Qwen3.5 streaming inference failure are not clearly the same bug." + }, + { + "left": "issue:44075", + "right": "issue:45656", + "accept": false, + "reason": "Optimizer args ignored and optimizer step called twice with deepspeed are different optimizer bugs." + }, + { + "left": "issue:43066", + "right": "issue:43122", + "accept": false, + "reason": "Wrong tokenizer decoder type and tokenization changes from 4.57.3 to 5.0 are related but not the same concrete defect." + }, + { + "left": "issue:42371", + "right": "issue:42831", + "accept": false, + "reason": "TF32 behavior settings and FineGrainedFP8 accuracy problems 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": 43, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:38617", + "issue:41628", + "issue:42222", + "issue:42757", + "issue:43012", + "issue:43064", + "issue:43097", + "issue:43335", + "issue:43408", + "issue:43441", + "issue:43452", + "issue:43502", + "issue:43531", + "issue:43577", + "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: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:45459", + "issue:45478", + "issue:45542", + "issue:45588", + "issue:46050", + "issue:46123" + ], + "soft_pairs": [ + "issue:43012|issue:43577", + "issue:43673|issue:45341", + "issue:44410|issue:45478", + "issue:44484|issue:45478", + "issue:43097|issue:44493", + "issue:44263|issue:44485", + "issue:41628|issue:45071", + "issue:43874|issue:44261", + "issue:43012|issue:43408", + "issue:42222|issue:45459", + "issue:45071|issue:45542", + "issue:43441|issue:44829", + "issue:43931|issue:44410", + "issue:43673|issue:46123", + "issue:43931|issue:44038", + "issue:43097|issue:43452", + "issue:45003|issue:45341", + "issue:43335|issue:45071", + "issue:43986|issue:45341", + "issue:44117|issue:44987", + "issue:43408|issue:44368", + "issue:43874|issue:44263", + "issue:43986|issue:44484", + "issue:43986|issue:45372", + "issue:43531|issue:45341", + "issue:38617|issue:42757", + "issue:44261|issue:44485", + "issue:43064|issue:44315", + "issue:43097|issue:46050", + "issue:43097|issue:45588", + "issue:38617|issue:44351", + "issue:44062|issue:44117", + "issue:44410|issue:45375", + "issue:44038|issue:44410", + "issue:43502|issue:43992", + "issue:44704|issue:45103", + "issue:43097|issue:43873", + "issue:43986|issue:44117", + "issue:43824|issue:43986" + ], + "prepared_review_unit_hash": "5664ec889ede3dfc1bf447ad1cbadd465b6c14769b328776690fc5e0c56d7d3d", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 39, + "serialized_chars": 39486, + "estimated_input_tokens": 9872, + "estimated_eval_tokens": 20000 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The soft edges are mostly superficial title-level similarities (imports, loading, config, flash-attention, Qwen/GLM families), but they point to different concrete failure modes and should not be merged as duplicates.", + "confidence": 0.79, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "issue:43502 is the strongest standalone issue in this batch: it states a concrete, user-visible invariant violation (local_files_only still triggering API requests) with a clear reproduction target and broad impact.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43012", + "right": "issue:43577", + "accept": false, + "reason": "Both mention dtype/precision, but one is a PyTorch warning during compilation and the other is an incorrect float32 load state in BLIP2; different code paths and symptoms." + }, + { + "left": "issue:43673", + "right": "issue:45341", + "accept": false, + "reason": "Completely unrelated areas: generation cache/chunked prefill versus a testing utility bug." + }, + { + "left": "issue:44410", + "right": "issue:45478", + "accept": false, + "reason": "Different models and failures: missing projections in qwen3next versus a from_pretrained error for Qwen3.5 MoE." + }, + { + "left": "issue:44484", + "right": "issue:45478", + "accept": false, + "reason": "Unrelated: a save_pretrained shard-size question versus a model loading error." + }, + { + "left": "issue:43097", + "right": "issue:44493", + "accept": false, + "reason": "Both are version/regression style reports, but one is about a removed config field and the other about unexpected position-id keys; not the same bug." + }, + { + "left": "issue:44263", + "right": "issue:44485", + "accept": false, + "reason": "Both are GLM-related, but they describe different concrete faults in different subsystems (torch.split behavior vs RoPE implementation)." + }, + { + "left": "issue:41628", + "right": "issue:45071", + "accept": false, + "reason": "Different surfaces entirely: an import failure for AutoImageProcessor versus a PretrainedConfig type-checking regression." + }, + { + "left": "issue:43874", + "right": "issue:44261", + "accept": false, + "reason": "Different bug classes: a missing image-processor method versus a numerical precision/config issue in MLA q_a_layernorm." + }, + { + "left": "issue:43012", + "right": "issue:43408", + "accept": false, + "reason": "One is a precision warning; the other is a model-type mismatch warning. Similar in being warnings, but not the same underlying issue." + }, + { + "left": "issue:42222", + "right": "issue:45459", + "accept": false, + "reason": "Unrelated: vitpose model breakage versus tokenizer protobuf error handling." + }, + { + "left": "issue:45071", + "right": "issue:45542", + "accept": false, + "reason": "Unrelated regression types: config type checking versus tensorboard/TensorFlow backend detection." + }, + { + "left": "issue:43441", + "right": "issue:44829", + "accept": false, + "reason": "Both involve flash attention, but one is a model-specific failure and the other is degenerate training under flash_attention_3; not the same code-path problem." + }, + { + "left": "issue:43931", + "right": "issue:44410", + "accept": false, + "reason": "Different model families and failure modes: weight-shape mismatch versus missing attention projections." + }, + { + "left": "issue:43673", + "right": "issue:46123", + "accept": false, + "reason": "Generation cache missing in chunked prefill versus a pipeline batching flag bug; different code paths." + }, + { + "left": "issue:43931", + "right": "issue:44038", + "accept": false, + "reason": "Both are Qwen3-VL related, but one is a shape-mismatch load failure and the other is a Qwen3-VL-Moe issue in transformers v5; too different to merge." + }, + { + "left": "issue:43097", + "right": "issue:43452", + "accept": false, + "reason": "Different loading problems: removed tie_embeddings handling versus gguf_file breaking tokenizer/model from_pretrained." + }, + { + "left": "issue:45003", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: sys.modules access in modeling_utils versus a testing_utils bug." + }, + { + "left": "issue:43335", + "right": "issue:45071", + "accept": false, + "reason": "Different areas and failure modes: SwitchTransformers config construction versus config type checking." + }, + { + "left": "issue:43986", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: video-model loading without torchvision versus a testing utility bug." + }, + { + "left": "issue:44117", + "right": "issue:44987", + "accept": false, + "reason": "Both are loading-related, but one is a tokenizer mapping assumption and the other is a model-specific transformers>=5.1.0 failure; not the same bug." + }, + { + "left": "issue:43408", + "right": "issue:44368", + "accept": false, + "reason": "Different models and issues: SAM3 model-type warning versus a tie_word_embeddings warning in Qwen3.5 LoRA fine-tuning." + }, + { + "left": "issue:43874", + "right": "issue:44263", + "accept": false, + "reason": "Both are GLM-adjacent, but one is an image-processor method omission and the other is a torch.split return-value issue; not the same defect." + }, + { + "left": "issue:43986", + "right": "issue:44484", + "accept": false, + "reason": "No overlap in bug shape: torchvision-dependent processor loading versus save_pretrained shard sizing." + }, + { + "left": "issue:43986", + "right": "issue:45372", + "accept": false, + "reason": "Both are processor/loading failures, but the root causes differ completely (missing torchvision versus missing ReasoningEffort import)." + }, + { + "left": "issue:43531", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated: Qwen3-MoE sliding-window behavior versus a testing utility bug." + }, + { + "left": "issue:38617", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols from different packages and are not the same underlying breakage." + }, + { + "left": "issue:44261", + "right": "issue:44485", + "accept": false, + "reason": "Both mention GLM/MLA-related math, but one is an epsilon/config precision issue and the other is a RoPE implementation bug." + }, + { + "left": "issue:43064", + "right": "issue:44315", + "accept": false, + "reason": "Different training-stack problems: wrong optimizer state placement under FSDP2/PEFT versus Liger Kernel not being applied with model_init." + }, + { + "left": "issue:43097", + "right": "issue:46050", + "accept": false, + "reason": "Different persistence/load regressions: removed tie_embeddings_and_encoder_decoder versus broken quantized save/load flow." + }, + { + "left": "issue:43097", + "right": "issue:45588", + "accept": false, + "reason": "Unrelated code paths: config-field removal versus a flash_attention crash on s_aux=None." + }, + { + "left": "issue:38617", + "right": "issue:44351", + "accept": false, + "reason": "Both are import-name errors, but they are different missing APIs and likely different release breakages." + }, + { + "left": "issue:44062", + "right": "issue:44117", + "accept": false, + "reason": "Tokenizer construction error versus tokenizer mapping returning None; different concrete problems." + }, + { + "left": "issue:44410", + "right": "issue:45375", + "accept": false, + "reason": "Different model/config fields and different failures; not the same bug." + }, + { + "left": "issue:44038", + "right": "issue:44410", + "accept": false, + "reason": "Related model family, but distinct failures: a Qwen3-VL-Moe loading issue versus missing projections in qwen3next." + }, + { + "left": "issue:43502", + "right": "issue:43992", + "accept": false, + "reason": "Different load-time issues: unwanted network access versus missing embed_tokens.weight in UMT5Encoder." + }, + { + "left": "issue:44704", + "right": "issue:45103", + "accept": false, + "reason": "Both involve kwargs handling, but one is AutoProcessor forwarding to cached_file and the other is an auto_docstring AttributeError under future annotations; not the same bug." + }, + { + "left": "issue:43097", + "right": "issue:43873", + "accept": false, + "reason": "Different areas: config-field removal versus offloading behavior under quantization." + }, + { + "left": "issue:43986", + "right": "issue:44117", + "accept": false, + "reason": "Different failure modes: missing torchvision during video-model loading versus TOKENIZER_MAPPING_NAMES returning None." + }, + { + "left": "issue:43824", + "right": "issue:43986", + "accept": false, + "reason": "Different models and causes: missing Qwen2_5_VLForConditionalGeneration import versus a video-model processor crash without torchvision." + } + ] + }, + "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": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:36683", + "issue:38617", + "issue:42490", + "issue:42757", + "issue:42907", + "issue:42994", + "issue:43012", + "issue:43064", + "issue:43066", + "issue:43208", + "issue:43299", + "issue:43352", + "issue:43475", + "issue:43502", + "issue:43526", + "issue:43550", + "issue:43582", + "issue:43618", + "issue:43756", + "issue:43761", + "issue:43792", + "issue:43824", + "issue:43825", + "issue:43867", + "issue:43901", + "issue:43906", + "issue:43976", + "issue:43992", + "issue:44038", + "issue:44222", + "issue:44230", + "issue:44368", + "issue:44380", + "issue:44483", + "issue:44484", + "issue:44485", + "issue:44514", + "issue:44655", + "issue:44683", + "issue:45276", + "issue:45290", + "issue:45381", + "issue:45478", + "issue:45479", + "issue:45753", + "issue:46077", + "issue:46123", + "issue:46132" + ], + "soft_pairs": [ + "issue:43066|issue:43906", + "issue:43208|issue:43550", + "issue:43992|issue:44484", + "issue:43825|issue:44655", + "issue:43012|issue:43550", + "issue:43012|issue:44368", + "issue:43352|issue:44380", + "issue:42907|issue:44655", + "issue:42907|issue:45753", + "issue:43792|issue:46123", + "issue:43352|issue:43976", + "issue:44380|issue:46132", + "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:42490|issue:43012", + "issue:42757|issue:43824", + "issue:43618|issue:43761", + "issue:43618|issue:44380", + "issue:38617|issue:43502", + "issue:44038|issue:44230", + "issue:44483|issue:45290", + "issue:44483|issue:45381", + "issue:43299|issue:43867", + "issue:43756|issue:44485", + "issue:36683|issue:43475", + "issue:43299|issue:45753", + "issue:42994|issue:44655", + "issue:43901|issue:44655", + "issue:45276|issue:46077" + ], + "prepared_review_unit_hash": "709fd6eb04bf27f0abe7c58af973776b3f587d68b367d06883265a01bfaa1fe5", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 34, + "serialized_chars": 37656, + "estimated_input_tokens": 9414, + "estimated_eval_tokens": 19084 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The set is very heterogeneous: most items are unrelated model-specific regressions, loading/saving issues, or API/behavior changes. None of the soft pairs look like true duplicates, so I rejected them all. If one issue must represent the cluster, the broad v5 chat-completions regression is the closest fit, but it does not explain the rest of the items.", + "confidence": 0.37, + "canonical_issue_reason": "issue:44483 is the broadest user-facing regression in the set and best matches the loose v5 API-breakage theme, though the cluster is too mixed for it to be a strong canonical duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:44483 is the most representative single issue by scope and severity, but only as a loose umbrella for the cluster\u2014not as a duplicate of the others.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43066", + "right": "issue:43906", + "accept": false, + "reason": "Different bugs: tokenizer decoder type vs an isolated reproduction for a separate issue." + }, + { + "left": "issue:43208", + "right": "issue:43550", + "accept": false, + "reason": "Unrelated model/training bugs; xLSTM training failures are not the same as Bamba SDPA compile failures." + }, + { + "left": "issue:43992", + "right": "issue:44484", + "accept": false, + "reason": "Missing embed_tokens.weight on load is not the same as a save_pretrained max_shard_size question." + }, + { + "left": "issue:43825", + "right": "issue:44655", + "accept": false, + "reason": "One is a pipeline task support/error-message issue; the other is pipeline serialization. Different code paths." + }, + { + "left": "issue:43012", + "right": "issue:43550", + "accept": false, + "reason": "A bfloat16 warning during compilation is not the same bug as Bamba torch.compile with SDPA." + }, + { + "left": "issue:43012", + "right": "issue:44368", + "accept": false, + "reason": "Both mention warnings, but they concern different configs and different behaviors." + }, + { + "left": "issue:43352", + "right": "issue:44380", + "accept": false, + "reason": "Different concrete failures: Flash Attention support gating vs GPT2 attention scaling being ignored under SDPA/FlashAttention." + }, + { + "left": "issue:42907", + "right": "issue:44655", + "accept": false, + "reason": "Both involve saving, but one is quantized/dequantized model saving and the other is saving Pipeline objects." + }, + { + "left": "issue:42907", + "right": "issue:45753", + "accept": false, + "reason": "Completely different symptoms: model saving vs generation looping on a specific image." + }, + { + "left": "issue:43792", + "right": "issue:46123", + "accept": false, + "reason": "Whisper model loading/running issue vs MaskGenerationPipeline partial-batch result dropping." + }, + { + "left": "issue:43352", + "right": "issue:43976", + "accept": false, + "reason": "Flash Attention support in a model is unrelated to Python version compatibility." + }, + { + "left": "issue:44380", + "right": "issue:46132", + "accept": false, + "reason": "Attention scaling being ignored is not the same as AttentionInterface.register changing function behavior." + }, + { + "left": "issue:44380", + "right": "issue:45479", + "accept": false, + "reason": "Attention backend bug vs sequence-classification zero-loss bug; different model behavior and code path." + }, + { + "left": "issue:43064", + "right": "issue:44380", + "accept": false, + "reason": "Distributed training optimizer-state corruption is unrelated to GPT2 attention scaling under SDPA/FlashAttention." + }, + { + "left": "issue:44222", + "right": "issue:45478", + "accept": false, + "reason": "Both are Qwen3.5/MoE-adjacent, but one is FP8 save_pretrained and the other is from_pretrained loading failure." + }, + { + "left": "issue:44483", + "right": "issue:44514", + "accept": false, + "reason": "Chat-completions request rejection is not the same as a Qwen2_5_VLProcessor batching crash." + }, + { + "left": "issue:36683", + "right": "issue:43582", + "accept": false, + "reason": "Gemma3Config missing vocab_size and Apple Silicon allocator warmup TypeError are unrelated." + }, + { + "left": "issue:43208", + "right": "issue:43526", + "accept": false, + "reason": "xLSTM training bugs and BeitImageProcessorFast label reduction are different subsystems and failures." + }, + { + "left": "issue:43012", + "right": "issue:44683", + "accept": false, + "reason": "Precision warning in compilation is not the same as compiled flex_attention failing on torch>=2.9." + }, + { + "left": "issue:42490", + "right": "issue:43012", + "accept": false, + "reason": "Path-with-slash import warning and bfloat16 compilation warning are unrelated." + }, + { + "left": "issue:42757", + "right": "issue:43824", + "accept": false, + "reason": "Both are import errors, but they concern different symbols and packages." + }, + { + "left": "issue:43618", + "right": "issue:43761", + "accept": false, + "reason": "CLIP attentions being unset is different from CLIPVisionModel hidden_states regression." + }, + { + "left": "issue:43618", + "right": "issue:44380", + "accept": false, + "reason": "Different attention-related bugs with different models and failure modes." + }, + { + "left": "issue:38617", + "right": "issue:43502", + "accept": false, + "reason": "ImportError from transformers.configuration_utils is unrelated to local_files_only still allowing API requests." + }, + { + "left": "issue:44038", + "right": "issue:44230", + "accept": false, + "reason": "Both mention Qwen3-VL-Moe/FP8, but one is a load/regression issue and the other is FP8 inference support." + }, + { + "left": "issue:44483", + "right": "issue:45290", + "accept": false, + "reason": "v1/chat/completions rejection is not the same as apply_chat_template crashing on tool-call assistant messages." + }, + { + "left": "issue:44483", + "right": "issue:45381", + "accept": false, + "reason": "Different layers: chat-completions request handling vs Qwen2.5-VL video position IDs." + }, + { + "left": "issue:43299", + "right": "issue:43867", + "accept": false, + "reason": "Qwen3VL-Moe loading break and sorted-state_dict load error are distinct load-path bugs." + }, + { + "left": "issue:43756", + "right": "issue:44485", + "accept": false, + "reason": "Smollm3 RoPE-layer loss and GLM-5 RoPE implementation discussion are not the same concrete bug." + }, + { + "left": "issue:36683", + "right": "issue:43475", + "accept": false, + "reason": "Missing Gemma3Config vocab_size and missing Sam3VisionEncoderOutput field are separate attribute mismatches." + }, + { + "left": "issue:43299", + "right": "issue:45753", + "accept": false, + "reason": "Model loading regression vs infinite loop on a specific image; unrelated symptoms." + }, + { + "left": "issue:42994", + "right": "issue:44655", + "accept": false, + "reason": "Quantized model saving failure is not the same as pipeline save_pretrained serialization." + }, + { + "left": "issue:43901", + "right": "issue:44655", + "accept": false, + "reason": "Docs mismatch for return_all_scores is unrelated to pipeline serialization failure." + }, + { + "left": "issue:45276", + "right": "issue:46077", + "accept": false, + "reason": "Gemma4 embedding resize behavior and bidirectional-attention mask construction are different bugs in the same family, not the same issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 45, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:36010", + "issue:36296", + "issue:36683", + "issue:41669", + "issue:42490", + "issue:42757", + "issue:42890", + "issue:42898", + "issue:43010", + "issue:43064", + "issue:43066", + "issue:43352", + "issue:43525", + "issue:43575", + "issue:43611", + "issue:43618", + "issue:43867", + "issue:43931", + "issue:43937", + "issue:43976", + "issue:43992", + "issue:44075", + "issue:44077", + "issue:44242", + "issue:44246", + "issue:44263", + "issue:44355", + "issue:44380", + "issue:44492", + "issue:44625", + "issue:44655", + "issue:44855", + "issue:44938", + "issue:45071", + "issue:45216", + "issue:45278", + "issue:45310", + "issue:45357", + "issue:45375", + "issue:45381", + "issue:45440", + "issue:45478", + "issue:45561", + "issue:45593", + "issue:45850", + "issue:46077", + "issue:46082" + ], + "soft_pairs": [ + "issue:36010|issue:42757", + "issue:42898|issue:43066", + "issue:43352|issue:43937", + "issue:44077|issue:44380", + "issue:36683|issue:44077", + "issue:43867|issue:45216", + "issue:43611|issue:44655", + "issue:43976|issue:44938", + "issue:44625|issue:46077", + "issue:46077|issue:46082", + "issue:43352|issue:45593", + "issue:36683|issue:43525", + "issue:43618|issue:45216", + "issue:44263|issue:45440", + "issue:36683|issue:44242", + "issue:43352|issue:43931", + "issue:43064|issue:44492", + "issue:43010|issue:45278", + "issue:43992|issue:45310", + "issue:43867|issue:45375", + "issue:44355|issue:44855", + "issue:41669|issue:44246", + "issue:36683|issue:43064", + "issue:42890|issue:45561", + "issue:44938|issue:45071", + "issue:43867|issue:45357", + "issue:43010|issue:45850", + "issue:43992|issue:45478", + "issue:43010|issue:43976", + "issue:44938|issue:45381", + "issue:43618|issue:44077", + "issue:42490|issue:44075", + "issue:36296|issue:43575" + ], + "prepared_review_unit_hash": "bf12d1f4bc0b29649ffdd852c438d910f54fcec924efebcaa5529a53a399de7c", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 47, + "item_count": 47, + "soft_pair_count": 33, + "serialized_chars": 36528, + "estimated_input_tokens": 9132, + "estimated_eval_tokens": 18520 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a heterogeneous mix of unrelated Transformers bug reports. The soft pairs share broad themes like import regressions, config serialization, or model-loading issues, but they do not look like the same concrete underlying bug. If one issue must represent the cluster, the Qwen3.5 save/load regression is the most concrete and actionable, but it is not a true duplicate hub for the rest.", + "confidence": 0.77, + "canonical_issue_reason": "Issue 45216 is the most concrete regression-style report with a specific model, API, and failure mode (`save_pretrained` on Qwen3.5). It is better triage material than the broader meta/issues, though it does not unify the whole cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45216 is the best single representative because it is narrowly scoped, reproducible, and framed as a regression with clear user impact. The other items are either broader, older, or about different models/subsystems.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36010", + "right": "issue:42757", + "accept": false, + "reason": "Both are import errors, but they involve different missing symbols/modules and likely different root causes." + }, + { + "left": "issue:42898", + "right": "issue:43066", + "accept": false, + "reason": "One is about `clean_up_tokenization_spaces` behavior, the other about tokenizer decoder type; same area, different bug." + }, + { + "left": "issue:43352", + "right": "issue:43937", + "accept": false, + "reason": "Flash Attention support gating and invalid `GenerationConfig` are unrelated failure modes." + }, + { + "left": "issue:44077", + "right": "issue:44380", + "accept": false, + "reason": "These touch different code paths: optimizer argument handling versus GPT-2 attention scaling under SDPA/FlashAttention." + }, + { + "left": "issue:36683", + "right": "issue:44077", + "accept": false, + "reason": "Missing `vocab_size` in Gemma3Config and `patchtsmixer` post-init restrictions are unrelated config issues." + }, + { + "left": "issue:43867", + "right": "issue:45216", + "accept": false, + "reason": "Both concern model loading/saving, but one is a generic sorted-state_dict load failure and the other is a Qwen3.5 save_pretrained regression." + }, + { + "left": "issue:43611", + "right": "issue:44655", + "accept": false, + "reason": "`base_model_prefix` loading failures and pipeline serialization are different save/load paths with different symptoms." + }, + { + "left": "issue:43976", + "right": "issue:44938", + "accept": false, + "reason": "Python version compatibility problems, but one targets 3.10+ behavior and the other 3.14 load failure; not the same bug." + }, + { + "left": "issue:44625", + "right": "issue:46077", + "accept": false, + "reason": "Qwen3.5 `num_labels` propagation and Gemma4 bidirectional attention masking are unrelated model-config bugs." + }, + { + "left": "issue:46077", + "right": "issue:46082", + "accept": false, + "reason": "One is an attention-mask construction bug, the other is `head_dim` validation in LlamaConfig; different issues." + }, + { + "left": "issue:43352", + "right": "issue:45593", + "accept": false, + "reason": "Flash Attention support and D-FINE auxiliary losses are entirely separate model behaviors." + }, + { + "left": "issue:36683", + "right": "issue:43525", + "accept": false, + "reason": "Both are missing-config-attribute errors, but for different models and different fields; not the same underlying bug." + }, + { + "left": "issue:43618", + "right": "issue:45216", + "accept": false, + "reason": "CLIPOutput attentions regression and Qwen3.5 save_pretrained regression are unrelated." + }, + { + "left": "issue:44263", + "right": "issue:45440", + "accept": false, + "reason": "`torch.split()` return-value handling and DeepSeekV3 MoE divergence are different implementation problems." + }, + { + "left": "issue:36683", + "right": "issue:44242", + "accept": false, + "reason": "A missing config attribute and an MoE load-balancing loss bug are unrelated." + }, + { + "left": "issue:43352", + "right": "issue:43931", + "accept": false, + "reason": "Flash Attention support error and Qwen3-VL weight-shape mismatch are different model-loading failures." + }, + { + "left": "issue:43064", + "right": "issue:44492", + "accept": false, + "reason": "FSDP2 optimizer state corruption and a typo in cache strategy docs/code are not the same issue." + }, + { + "left": "issue:43010", + "right": "issue:45278", + "accept": false, + "reason": "`@torch.no_grad` on cache updates and broad import failures after upgrade are too different to merge." + }, + { + "left": "issue:43992", + "right": "issue:45310", + "accept": false, + "reason": "UMT5Encoder missing weights and Qwen3.5 MoE from_pretrained errors affect different models and code paths." + }, + { + "left": "issue:43867", + "right": "issue:45375", + "accept": false, + "reason": "Generic load-state_dict ordering failure versus missing Qwen3.5 vision config field; related serialization area, but not the same bug." + }, + { + "left": "issue:44355", + "right": "issue:44855", + "accept": false, + "reason": "Compiled-file import errors and a Python 3.13 `@torch.jit.script` parsing problem are different import-time failures." + }, + { + "left": "issue:41669", + "right": "issue:44246", + "accept": false, + "reason": "Import performance issues are broad, but one is about removing `import *` and the other about sporadic slow imports." + }, + { + "left": "issue:36683", + "right": "issue:43064", + "accept": false, + "reason": "Gemma3 config attribute errors and FSDP2 optimizer-state corruption are unrelated." + }, + { + "left": "issue:42890", + "right": "issue:45561", + "accept": false, + "reason": "Both are test instability issues, but they involve different races and different files; not one underlying bug." + }, + { + "left": "issue:44938", + "right": "issue:45071", + "accept": false, + "reason": "Python 3.14 load failure and `PretrainedConfig` type-checking regression are separate compatibility problems." + }, + { + "left": "issue:43867", + "right": "issue:45357", + "accept": false, + "reason": "Both touch loading/saving, but one is a generic state_dict ordering issue and the other is a specific incorrect visual encoder key regression." + }, + { + "left": "issue:43010", + "right": "issue:45850", + "accept": false, + "reason": "A concrete cache-update no-grad fix and a broad complaint that v5 breaks many models are not duplicates." + }, + { + "left": "issue:43992", + "right": "issue:45478", + "accept": false, + "reason": "Different models, different missing/misconfigured weights, and different from_pretrained failures." + }, + { + "left": "issue:43010", + "right": "issue:43976", + "accept": false, + "reason": "Cache mutation under no-grad and Python version support are unrelated." + }, + { + "left": "issue:44938", + "right": "issue:45381", + "accept": false, + "reason": "One is a Python 3.14 load failure; the other is a Qwen2.5-VL video position-id bug." + }, + { + "left": "issue:43618", + "right": "issue:44077", + "accept": false, + "reason": "CLIP attentions missing and `patchtsmixer` post_init policy are unrelated model regressions." + }, + { + "left": "issue:42490", + "right": "issue:44075", + "accept": false, + "reason": "A trailing-slash local-model import warning and unused SGD args are unrelated." + }, + { + "left": "issue:36296", + "right": "issue:43575", + "accept": false, + "reason": "Both mention tensor parallelism, but one is a training bug and the other is an OOM on model load; not the same 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": 46, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43576", + "issue:43618", + "issue:43723", + "issue:43827", + "issue:43867", + "issue:43874", + "issue:44062", + "issue:44112", + "issue:44222", + "issue:44230", + "issue:44315", + "issue:44351", + "issue:44355", + "issue:44360", + "issue:44387", + "issue:44484", + "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:28282|issue:33453", + "issue:44360|issue:45440", + "issue:43867|issue:44484", + "issue:43618|issue:44857", + "issue:43618|issue:44977", + "issue:42890|issue:44964", + "issue:42890|issue:44112", + "issue:42890|issue:45600", + "issue:43502|issue:44062", + "issue:43723|issue:44556", + "issue:42548|issue:44351", + "issue:43576|issue:46132", + "issue:42890|issue:45341", + "issue:43010|issue:43065", + "issue:44387|issue:44556", + "issue:44222|issue:45125", + "issue:43502|issue:44485", + "issue:44355|issue:44749", + "issue:45125|issue:45478", + "issue:43010|issue:43232", + "issue:30990|issue:31795", + "issue:30333|issue:33453", + "issue:45125|issue:45375", + "issue:43502|issue:44222", + "issue:43299|issue:44556", + "issue:43502|issue:44961", + "issue:44315|issue:44360", + "issue:42831|issue:43874", + "issue:44355|issue:45071", + "issue:30333|issue:36296", + "issue:43827|issue:44961", + "issue:44355|issue:44655", + "issue:44222|issue:44230", + "issue:33453|issue:34689", + "issue:42757|issue:43576", + "issue:33453|issue:36296", + "issue:43576|issue:45071" + ], + "prepared_review_unit_hash": "d9fb1f837bf1757ab9f3863cbbba6de589ccf0b9870a119f56196a9b862a0c26", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 37, + "serialized_chars": 38147, + "estimated_input_tokens": 9537, + "estimated_eval_tokens": 19330 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is highly heterogeneous: most items are unrelated bugs, docs, CI, or feature regressions. The only mild theme is \u201cloading/import regressions,\u201d but none of the soft pairs look like the same concrete bug or change, so I would not merge any of them. If forced to anchor the cluster, #33453 is the broadest tokenizer-loading regression.", + "confidence": 0.74, + "canonical_issue_reason": "#33453 is the broadest and most reusable anchor among the loading-related reports: it names a general tokenizer-loading regression, whereas most other candidates are narrower or point to different subsystems. It is still only a loose anchor because the cluster is not actually a duplicate set.", + "canonical_pr_reason": null, + "best_issue_reason": "#33453 is the best single issue to represent the small loading-regression subset because it is broad, explicit, and closer in scope to the other tokenizer/model-loading complaints than the more specialized reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:28282", + "right": "issue:33453", + "accept": false, + "reason": "PyTorch import error vs tokenizer loading regression; different failure mode and code path." + }, + { + "left": "issue:44360", + "right": "issue:45440", + "accept": false, + "reason": "DSA indexer/ReLU discussion is unrelated to DeepSeekV3 MoE divergence." + }, + { + "left": "issue:43867", + "right": "issue:44484", + "accept": false, + "reason": "State-dict sorting load bug vs save_pretrained shard-size question; unrelated." + }, + { + "left": "issue:43618", + "right": "issue:44857", + "accept": false, + "reason": "CLIP attentions regression is unrelated to LwDetrImageLoss AMP crash." + }, + { + "left": "issue:43618", + "right": "issue:44977", + "accept": false, + "reason": "CLIPOutput attentions issue and Qwen flash-attention generation bug are different code paths." + }, + { + "left": "issue:42890", + "right": "issue:44964", + "accept": false, + "reason": "SamHQ test seeding flakiness vs Phi-4 multimodal load failure; not the same bug." + }, + { + "left": "issue:42890", + "right": "issue:44112", + "accept": false, + "reason": "Both are test-related, but one is seed stability and the other is a stale device override CI failure." + }, + { + "left": "issue:42890", + "right": "issue:45600", + "accept": false, + "reason": "Test seeding issue vs auto_mappings config reference breakage; unrelated." + }, + { + "left": "issue:43502", + "right": "issue:44062", + "accept": false, + "reason": "local_files_only network leak vs AddedToken kwarg error; different behaviors." + }, + { + "left": "issue:43723", + "right": "issue:44556", + "accept": false, + "reason": "Tokenizer loading regression and checkpoint reload regression are both loading-related, but not the same concrete failure." + }, + { + "left": "issue:42548", + "right": "issue:44351", + "accept": false, + "reason": "Different missing exports: PreTrainedModel vs HybridCache." + }, + { + "left": "issue:43576", + "right": "issue:46132", + "accept": false, + "reason": "Broken CLI env command vs changed register behavior; unrelated." + }, + { + "left": "issue:42890", + "right": "issue:45341", + "accept": false, + "reason": "SamHQ seed flakiness vs generic testing_utils bug; not the same issue." + }, + { + "left": "issue:43010", + "right": "issue:43065", + "accept": false, + "reason": "Cache/Layers no_grad decoration request vs Sam3PixelDecoder dummy Conv2d; different subsystems." + }, + { + "left": "issue:44387", + "right": "issue:44556", + "accept": false, + "reason": "Int4 CUDA memory regression vs checkpoint reload after upgrade; different bugs." + }, + { + "left": "issue:44222", + "right": "issue:45125", + "accept": false, + "reason": "FP8 save_pretrained bug vs missing tensor-parallel plan; same general area, but not one fixable PR." + }, + { + "left": "issue:43502", + "right": "issue:44485", + "accept": false, + "reason": "local_files_only network leak vs RoPE implementation discussion; unrelated." + }, + { + "left": "issue:44355", + "right": "issue:44749", + "accept": false, + "reason": "Compiled Python file errors vs performance slowdown after upgrade; not the same failure." + }, + { + "left": "issue:45125", + "right": "issue:45478", + "accept": false, + "reason": "Missing _tp_plan and from_pretrained error on Qwen3.5 MoE may be related, but the reports describe different concrete breakages." + }, + { + "left": "issue:43010", + "right": "issue:43232", + "accept": false, + "reason": "No-grad decoration request vs generation kwargs update bug; different code paths." + }, + { + "left": "issue:30990", + "right": "issue:31795", + "accept": false, + "reason": "Sentence Transformers loading hang vs documentation confusion for model.forward." + }, + { + "left": "issue:30333", + "right": "issue:33453", + "accept": false, + "reason": "MLflow reporting on failed jobs is unrelated to tokenizer loading regressions." + }, + { + "left": "issue:45125", + "right": "issue:45375", + "accept": false, + "reason": "Tensor-parallel metadata issue vs missing vision config field; related model family, different bug." + }, + { + "left": "issue:43502", + "right": "issue:44222", + "accept": false, + "reason": "local_files_only network leak vs FP8 save_pretrained moe; different defects." + }, + { + "left": "issue:43299", + "right": "issue:44556", + "accept": false, + "reason": "Qwen3VL MoE load break and version-upgrade checkpoint reload failure are both loading issues, but not the same concrete regression." + }, + { + "left": "issue:43502", + "right": "issue:44961", + "accept": false, + "reason": "API network leak vs a junk/non-informative issue title; no overlap." + }, + { + "left": "issue:44315", + "right": "issue:44360", + "accept": false, + "reason": "Liger Kernel/model_init issue is unrelated to the DSA indexer ReLU discussion." + }, + { + "left": "issue:42831", + "right": "issue:43874", + "accept": false, + "reason": "FineGrainedFP8 accuracy issue vs GLM46V image processor attribute error; unrelated." + }, + { + "left": "issue:44355", + "right": "issue:45071", + "accept": false, + "reason": "Compiled-file runtime errors vs PretrainedConfig type checking regression; different bugs." + }, + { + "left": "issue:30333", + "right": "issue:36296", + "accept": false, + "reason": "MLflow reporting bug vs tensor-parallel training bug; unrelated." + }, + { + "left": "issue:43827", + "right": "issue:44961", + "accept": false, + "reason": "Docs still referencing pipeline() vs unrelated junk title; no shared bug." + }, + { + "left": "issue:44355", + "right": "issue:44655", + "accept": false, + "reason": "Compiled file execution errors vs inability to save Pipeline objects; different issues." + }, + { + "left": "issue:44222", + "right": "issue:44230", + "accept": false, + "reason": "FP8 save_pretrained bug vs FP8 inference support request; same broad theme, but distinct changes." + }, + { + "left": "issue:33453", + "right": "issue:34689", + "accept": false, + "reason": "Tokenizer loading regression vs Llama 3.2 vision model-loading breakage; different concrete failures." + }, + { + "left": "issue:42757", + "right": "issue:43576", + "accept": false, + "reason": "Missing import from huggingface_hub vs broken transformers env CLI; unrelated." + }, + { + "left": "issue:33453", + "right": "issue:36296", + "accept": false, + "reason": "Tokenizer loading regression vs tensor parallel training bug; unrelated." + }, + { + "left": "issue:43576", + "right": "issue:45071", + "accept": false, + "reason": "Broken env command vs PretrainedConfig type-checking regression; different code paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 1, + "component_count": 17, + "review_unit_index": 47, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "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:43824", + "issue:43901", + "issue:43976", + "issue:44038", + "issue:44230", + "issue:44246", + "issue:44261", + "issue:44273", + "issue:44295", + "issue:44297", + "issue:44315", + "issue:44355", + "issue:44360", + "issue:44393", + "issue:44485", + "issue:44492", + "issue:44556", + "issue:44623", + "issue:44704", + "issue:44908", + "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" + ], + "soft_pairs": [ + "issue:43576|issue:43824", + "issue:31795|issue:38617", + "issue:41628|issue:42548", + "issue:44961|issue:45468", + "issue:38617|issue:43576", + "issue:44908|issue:44961", + "issue:44485|issue:44961", + "issue:43673|issue:44492", + "issue:41628|issue:44355", + "issue:36296|issue:43901", + "issue:36296|issue:43526", + "issue:45362|issue:45850", + "issue:44261|issue:44360", + "issue:44623|issue:45230", + "issue:44273|issue:45542", + "issue:44273|issue:45003", + "issue:44038|issue:45125", + "issue:44393|issue:45071", + "issue:44393|issue:45753", + "issue:43576|issue:44556", + "issue:44230|issue:46095", + "issue:43525|issue:45850", + "issue:41669|issue:43673", + "issue:44556|issue:45092", + "issue:43572|issue:45850", + "issue:44492|issue:45850", + "issue:42371|issue:44492", + "issue:45278|issue:45850", + "issue:42890|issue:44908", + "issue:44704|issue:44908", + "issue:44246|issue:44273", + "issue:43408|issue:44315", + "issue:43976|issue:44393", + "issue:43352|issue:44393", + "issue:43576|issue:44908", + "issue:44297|issue:45230", + "issue:44908|issue:45071", + "issue:44295|issue:45230" + ], + "prepared_review_unit_hash": "1836fbe716b3a5430f10e94cfbbb67b2be32983e97c5441e14974148c4ac3d98", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 38, + "serialized_chars": 38439, + "estimated_input_tokens": 9610, + "estimated_eval_tokens": 19476 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a loose mix of Transformers v5 regression reports, especially import/loading breakages; only one soft pair looks like the same underlying compatibility bug.", + "confidence": 0.73, + "canonical_issue_reason": "issue:45278 is the clearest umbrella report for the repeated post-upgrade import/regression theme in this cluster, while still being concrete and actionable.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45278 is the best representative issue for deduping because it captures the broad import-breakage symptom seen across several nearby reports without being as vague as the generic v5 complaints.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43576", + "right": "issue:43824", + "accept": false, + "reason": "Different symptoms: broken env command vs a missing model export import error." + }, + { + "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:41628", + "right": "issue:42548", + "accept": false, + "reason": "Both are import-name failures, but they name different missing symbols and likely need separate export fixes." + }, + { + "left": "issue:44961", + "right": "issue:45468", + "accept": false, + "reason": "No meaningful overlap; one is a nonsense title, the other is a Gemma audio positional-encoding bug." + }, + { + "left": "issue:38617", + "right": "issue:43576", + "accept": false, + "reason": "ImportError for a missing helper is unrelated to the v5 env command regression." + }, + { + "left": "issue:44908", + "right": "issue:44961", + "accept": false, + "reason": "Scheduler kwargs bug and unrelated placeholder issue are not the same bug." + }, + { + "left": "issue:44485", + "right": "issue:44961", + "accept": false, + "reason": "GLM RoPE implementation discussion is unrelated to the placeholder issue." + }, + { + "left": "issue:43673", + "right": "issue:44492", + "accept": false, + "reason": "Chunked-prefill cache regression is unrelated to a typo in cache strategy docs/text." + }, + { + "left": "issue:41628", + "right": "issue:44355", + "accept": false, + "reason": "ImportError for AutoImageProcessor is unrelated to compiled-file execution errors." + }, + { + "left": "issue:36296", + "right": "issue:43901", + "accept": false, + "reason": "Tensor parallel training bug is unrelated to TextClassificationPipeline docs behavior." + }, + { + "left": "issue:36296", + "right": "issue:43526", + "accept": false, + "reason": "Tensor parallel training issue is unrelated to BeitImageProcessorFast label reduction." + }, + { + "left": "issue:45362", + "right": "issue:45850", + "accept": false, + "reason": "Specific chat crash vs broad 'v5 breaks too many models' complaint; not the same concrete bug." + }, + { + "left": "issue:44261", + "right": "issue:44360", + "accept": false, + "reason": "MLA layernorm precision issue is unrelated to the DSA ReLU discussion." + }, + { + "left": "issue:44623", + "right": "issue:45230", + "accept": false, + "reason": "Missing processor save files is too specific to match the generic bug report." + }, + { + "left": "issue:44273", + "right": "issue:45542", + "accept": false, + "reason": "Lazy loading failure and undefined tf backend error are different code paths." + }, + { + "left": "issue:44273", + "right": "issue:45003", + "accept": false, + "reason": "Lazy loading behavior and unsafe sys.modules access are related broadly, but not the same bug." + }, + { + "left": "issue:44038", + "right": "issue:45125", + "accept": false, + "reason": "Qwen3-VL/Moe bug and missing tensor-parallel plan are different issues." + }, + { + "left": "issue:44393", + "right": "issue:45071", + "accept": false, + "reason": "2D bounding-box hallucination is unrelated to PretrainedConfig type-checking breakage." + }, + { + "left": "issue:44393", + "right": "issue:45753", + "accept": false, + "reason": "Both are Qwen3-VL-family reports, but one is bbox output error and the other is an infinite loop on a specific image." + }, + { + "left": "issue:43576", + "right": "issue:44556", + "accept": false, + "reason": "Broken env command and checkpoint reload failure are both v5 regressions but not the same code-path." + }, + { + "left": "issue:44230", + "right": "issue:46095", + "accept": false, + "reason": "FP8 support request and deepseekv4 weight-conversion question are not duplicates." + }, + { + "left": "issue:43525", + "right": "issue:45850", + "accept": false, + "reason": "Missing pad_token_id in one config is not the same as a generic 'v5 breaks many models' complaint." + }, + { + "left": "issue:41669", + "right": "issue:43673", + "accept": false, + "reason": "Import-star performance regression is unrelated to the chunked_prefill cache bug." + }, + { + "left": "issue:44556", + "right": "issue:45092", + "accept": true, + "reason": "Both describe Transformers v5 compatibility regressions when loading older checkpoints, specifically around meta initialization/backward compatibility." + }, + { + "left": "issue:43572", + "right": "issue:45850", + "accept": false, + "reason": "StableLmConfig pad_token_idx regression is a concrete config bug, not the same as a broad v5 breakage report." + }, + { + "left": "issue:44492", + "right": "issue:45850", + "accept": false, + "reason": "A typo in cache strategies is not the same as a generic 'many models break' complaint." + }, + { + "left": "issue:42371", + "right": "issue:44492", + "accept": false, + "reason": "TF32 API settings guidance is unrelated to a cache-strategy typo." + }, + { + "left": "issue:45278", + "right": "issue:45850", + "accept": false, + "reason": "Both are broad v5 complaints, but one is specifically import errors after upgrade while the other is too vague to treat as the same bug." + }, + { + "left": "issue:42890", + "right": "issue:44908", + "accept": false, + "reason": "Test flakiness from missing seeds is unrelated to scheduler kwargs handling." + }, + { + "left": "issue:44704", + "right": "issue:44908", + "accept": false, + "reason": "AutoProcessor cached_file kwargs forwarding is unrelated to inverse_sqrt scheduler behavior." + }, + { + "left": "issue:44246", + "right": "issue:44273", + "accept": false, + "reason": "Import latency is a performance issue; lazy loading not working is a functional regression." + }, + { + "left": "issue:43408", + "right": "issue:44315", + "accept": false, + "reason": "sam3 model-type warning is unrelated to Liger Kernel application in model_init." + }, + { + "left": "issue:43976", + "right": "issue:44393", + "accept": false, + "reason": "Python version compatibility issue is unrelated to Qwen3-VL bbox output problems." + }, + { + "left": "issue:43352", + "right": "issue:44393", + "accept": false, + "reason": "Flash Attention 2 support error and Qwen3-VL bbox hallucination are different model bugs." + }, + { + "left": "issue:43576", + "right": "issue:44908", + "accept": false, + "reason": "Broken env command and inverse_sqrt scheduler kwargs are unrelated regressions." + }, + { + "left": "issue:44297", + "right": "issue:45230", + "accept": false, + "reason": "Tokenizer config class mismatch is unrelated to the generic bug report." + }, + { + "left": "issue:44908", + "right": "issue:45071", + "accept": false, + "reason": "Scheduler kwargs bug is unrelated to type-checking breakage in PretrainedConfig." + }, + { + "left": "issue:44295", + "right": "issue:45230", + "accept": false, + "reason": "Position_ids buffer access error is unrelated to the generic bug report." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: the cluster is mostly a mix of v5-era regressions, with import/loading issues standing out, and the chosen representative issue is a reasonable umbrella. The soft-edge verdicts are conservative overall, with the one accepted pair (44556/45092) supported by the titles, and the rejections generally 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": 48, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:31515", + "issue:31795", + "issue:33453", + "issue:34689", + "issue:37428", + "issue:38617", + "issue:41669", + "issue:42371", + "issue:42913", + "issue:43066", + "issue:43298", + "issue:43502", + "issue:43519", + "issue:43525", + "issue:43572", + "issue:43576", + "issue:43606", + "issue:43618", + "issue:43704", + "issue:43827", + "issue:43874", + "issue:44077", + "issue:44230", + "issue:44261", + "issue:44263", + "issue:44273", + "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:45092", + "issue:45230", + "issue:45290", + "issue:45310", + "issue:45341", + "issue:45468", + "issue:45478", + "issue:45850", + "issue:46095" + ], + "soft_pairs": [ + "issue:41669|issue:44351", + "issue:41669|issue:45341", + "issue:44492|issue:44945", + "issue:43874|issue:45850", + "issue:44393|issue:45478", + "issue:44077|issue:44861", + "issue:43827|issue:44908", + "issue:31795|issue:33453", + "issue:41669|issue:44908", + "issue:44315|issue:44829", + "issue:43576|issue:44273", + "issue:44360|issue:45468", + "issue:44655|issue:44861", + "issue:45230|issue:45310", + "issue:44273|issue:44908", + "issue:43519|issue:44485", + "issue:43519|issue:44908", + "issue:41669|issue:43704", + "issue:31795|issue:34689", + "issue:31795|issue:37428", + "issue:44315|issue:45092", + "issue:44485|issue:45468", + "issue:43704|issue:44485", + "issue: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:44861|issue:45290", + "issue:44998|issue:45850", + "issue:43618|issue:44998", + "issue:44998|issue:45341", + "issue:43525|issue:44861", + "issue:43572|issue:44861", + "issue:31515|issue:38617", + "issue:44230|issue:44623" + ], + "prepared_review_unit_hash": "8ee384339b0b393bee6350023486558cad32cdd291ab2b6992a46875ab5c909d", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 48, + "item_count": 48, + "soft_pair_count": 42, + "serialized_chars": 39663, + "estimated_input_tokens": 9916, + "estimated_eval_tokens": 20088 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly heterogeneous: a few items share broad themes like imports, loading, or tokenizer/model regressions, but none of the soft pairs look like the same concrete bug/change. All soft edges are rejected. Issue 41669 is the broadest and most central representative among the set.", + "confidence": 0.23, + "canonical_issue_reason": "issue:41669 is the broadest, most cross-cutting item here: a package-wide import cleanup/performance report with the most soft overlaps to other import-related issues.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:41669 is the best single anchor because it is the most general infrastructural issue and the one most plausibly related to multiple neighboring reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41669", + "right": "issue:44351", + "accept": false, + "reason": "Import-star cleanup/perf issue vs a specific missing HybridCache import/export error; same area, different bug." + }, + { + "left": "issue:41669", + "right": "issue:45341", + "accept": false, + "reason": "Broad import cleanup vs a small testing_utils bug; no shared failure mode." + }, + { + "left": "issue:44492", + "right": "issue:44945", + "accept": false, + "reason": "Typo in cache strategies vs incorrect pipeline-parallel LLM output; unrelated code paths." + }, + { + "left": "issue:43874", + "right": "issue:45850", + "accept": false, + "reason": "Specific GLM46V multimodal method bug vs a generic 'many models break' report; not the same concrete issue." + }, + { + "left": "issue:44393", + "right": "issue:45478", + "accept": false, + "reason": "Different Qwen3-VL/Gemma model-specific failures with different symptoms and fixes." + }, + { + "left": "issue:44077", + "right": "issue:44861", + "accept": false, + "reason": "PatchTSMixer post-init restriction vs tied-weight-key attribute crash; unrelated utilities." + }, + { + "left": "issue:43827", + "right": "issue:44908", + "accept": false, + "reason": "Docs referencing removed pipeline() API vs scheduler kwargs not being passed; docs issue vs runtime logic bug." + }, + { + "left": "issue:31795", + "right": "issue:33453", + "accept": false, + "reason": "Forward-doc confusion vs tokenizer-loading regression; both mention loading/docs, but not the same bug." + }, + { + "left": "issue:41669", + "right": "issue:44908", + "accept": false, + "reason": "Import-star cleanup vs scheduler configuration propagation; unrelated fixes." + }, + { + "left": "issue:44315", + "right": "issue:44829", + "accept": false, + "reason": "Liger-kernel application during model_init vs flash_attention_3 training degeneration; different mechanisms and symptoms." + }, + { + "left": "issue:43576", + "right": "issue:44273", + "accept": false, + "reason": "Broken env CLI vs lazy-loading malfunction; different features entirely." + }, + { + "left": "issue:44360", + "right": "issue:45468", + "accept": false, + "reason": "DSA indexer ReLU discussion vs Gemma-4 audio positional encoding bug; unrelated internals." + }, + { + "left": "issue:44655", + "right": "issue:44861", + "accept": false, + "reason": "save_pretrained() pipeline-saving issue vs tied-weight-key crash; save/export path vs weight-tie handling." + }, + { + "left": "issue:45230", + "right": "issue:45310", + "accept": false, + "reason": "Generic bug report placeholder vs a specific Qwen3.5 MoE from_pretrained regression; not duplicates." + }, + { + "left": "issue:44273", + "right": "issue:44908", + "accept": false, + "reason": "Lazy loading regression vs inverse_sqrt scheduler kwargs bug; unrelated subsystems." + }, + { + "left": "issue:43519", + "right": "issue:44485", + "accept": false, + "reason": "Qwen3VL timestamp calculation bug vs GLM-5 RoPE implementation discussion; different model families and code paths." + }, + { + "left": "issue:43519", + "right": "issue:44908", + "accept": false, + "reason": "Timestamp math bug vs scheduler kwargs handling; no common underlying change." + }, + { + "left": "issue:41669", + "right": "issue:43704", + "accept": false, + "reason": "Import cleanup/perf issue vs VRAM leak in Qwen3ForCausalLM threads; unrelated." + }, + { + "left": "issue:31795", + "right": "issue:34689", + "accept": false, + "reason": "Documentation confusion vs Llama 3.2 model loading regression; different failure modes." + }, + { + "left": "issue:31795", + "right": "issue:37428", + "accept": false, + "reason": "Forward-doc confusion vs missing flash-attention utility import; unrelated." + }, + { + "left": "issue:44315", + "right": "issue:45092", + "accept": false, + "reason": "Liger-kernel/model_init issue vs remote-code meta-init incompatibility for InternVL2; both init-related but different concrete problems." + }, + { + "left": "issue:44485", + "right": "issue:45468", + "accept": false, + "reason": "Both are RoPE/positional-encoding themed, but they target different models and implementations." + }, + { + "left": "issue:43704", + "right": "issue:44485", + "accept": false, + "reason": "VRAM leak under threaded dataloading vs RoPE implementation discussion; unrelated." + }, + { + "left": "issue:43298", + "right": "issue:46095", + "accept": false, + "reason": "Continuous batching audio-model support vs asking for a weight-conversion script; unrelated." + }, + { + "left": "issue:42371", + "right": "issue:43704", + "accept": false, + "reason": "TF32 settings guidance vs VRAM leak; unrelated topics." + }, + { + "left": "issue:43502", + "right": "issue:43519", + "accept": false, + "reason": "local_files_only network requests vs timestamp calculation bug; different code paths." + }, + { + "left": "issue:43519", + "right": "issue:44623", + "accept": false, + "reason": "Timestamp calculation bug vs processor.save_pretrained missing files; unrelated processor behaviors." + }, + { + "left": "issue:43519", + "right": "issue:43704", + "accept": false, + "reason": "Timestamp calculation bug vs threaded VRAM leak; no shared root cause." + }, + { + "left": "issue:43606", + "right": "issue:45468", + "accept": false, + "reason": "CPU offload device mismatch for bark-small vs Gemma-4 audio positional encoding; only loosely audio-related." + }, + { + "left": "issue:44263", + "right": "issue:45468", + "accept": false, + "reason": "GlmMoeDsaIndexer split-values issue vs Gemma-4 audio positional encoding bug; unrelated." + }, + { + "left": "issue:44261", + "right": "issue:45468", + "accept": false, + "reason": "MLA q_a_layernorm epsilon precision mismatch vs Gemma-4 audio positional encoding; different components." + }, + { + "left": "issue:42913", + "right": "issue:43066", + "accept": false, + "reason": "General tokenizer behavior drift from v4 to v5 vs wrong tokenizer decoder type in v5; related theme, but not the same concrete bug." + }, + { + "left": "issue:44908", + "right": "issue:44998", + "accept": false, + "reason": "Scheduler kwargs bug vs an unrelated placeholder issue title; no connection." + }, + { + "left": "issue:44492", + "right": "issue:44998", + "accept": false, + "reason": "Cache-strategy typo vs unrelated placeholder issue title." + }, + { + "left": "issue:44861", + "right": "issue:45290", + "accept": false, + "reason": "Tied-weight-key AttributeError vs chat template crash on tool-call assistant messages; different subsystems." + }, + { + "left": "issue:44998", + "right": "issue:45850", + "accept": false, + "reason": "Unrelated placeholder issue vs generic 'too many models break' complaint." + }, + { + "left": "issue:43618", + "right": "issue:44998", + "accept": false, + "reason": "CLIPOutput attentions regression vs unrelated placeholder issue." + }, + { + "left": "issue:44998", + "right": "issue:45341", + "accept": false, + "reason": "Unrelated placeholder issue vs small testing_utils bug." + }, + { + "left": "issue:43525", + "right": "issue:44861", + "accept": false, + "reason": "Llama4Config pad_token_id missing vs tied-weight-key crash; different model/config bugs." + }, + { + "left": "issue:43572", + "right": "issue:44861", + "accept": false, + "reason": "StableLmConfig pad_token_idx missing vs tied-weight-key crash; unrelated." + }, + { + "left": "issue:31515", + "right": "issue:38617", + "accept": false, + "reason": "Slow from_pretrained checkpoint loading vs missing layer_type_validation import; unrelated import/load issues." + }, + { + "left": "issue:44230", + "right": "issue:44623", + "accept": false, + "reason": "FP8 Qwen3-VL support request vs processor.save_pretrained missing files; feature request vs 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": 49, + "review_unit_count": 49, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-10000-413", + "nodes": [ + "issue:30333", + "issue:31515", + "issue:31795", + "issue:33453", + "issue:36296", + "issue:43298", + "issue:44230", + "issue:45310" + ], + "soft_pairs": [ + "issue:30333|issue:31515", + "issue:31515|issue:36296", + "issue:31515|issue:31795", + "issue:44230|issue:45310", + "issue:31515|issue:33453", + "issue:43298|issue:44230" + ], + "prepared_review_unit_hash": "25f2fb61eb2bd20385176d0c5a98b45c11dc4c910deeb14ab4d257c7d1037ebc", + "component_budget": { + "node_count": 413, + "item_count": 413, + "soft_pair_count": 1547, + "serialized_chars": 797880, + "estimated_input_tokens": 199470, + "estimated_eval_tokens": 399196 + }, + "budget": { + "node_count": 8, + "item_count": 8, + "soft_pair_count": 6, + "serialized_chars": 6307, + "estimated_input_tokens": 1577, + "estimated_eval_tokens": 3410 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "The cluster is heterogeneous: training/MLFlow reporting, checkpoint loading performance, documentation, tokenizer regression, tensor-parallel training, audio continuous batching, fp8 support, and a Qwen3.5 MoE load error. None of the soft pairs look like true duplicates. The most suitable issue to anchor the cluster is the concrete version-specific bug report in 45310.", + "confidence": 0.79, + "canonical_issue_reason": "issue:45310 is the clearest concrete bug report in the set: it names a specific failing path, model, and version regression. The other issues are either feature requests, docs, performance, or unrelated subsystems.", + "canonical_pr_reason": null, + "best_issue_reason": "issue:45310 is the most actionable and specific issue, with a direct bug/regression framing and a narrow reproduction surface. It is the strongest candidate to represent the cluster.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:30333", + "right": "issue:31515", + "accept": false, + "reason": "Both are about training/loading, but one is an MLFlow status/reporting failure after job failure and the other is slow checkpoint loading. Different code paths and different symptoms." + }, + { + "left": "issue:31515", + "right": "issue:36296", + "accept": false, + "reason": "Checkpoint-loading slowness is a performance problem; tensor parallel training bug is a separate training-parallelism issue. Not the same underlying defect." + }, + { + "left": "issue:31515", + "right": "issue:31795", + "accept": false, + "reason": "One is slow checkpoint loading; the other is confusing forward-docs for input_ids/past_key_values. Same broad area only at a very high level." + }, + { + "left": "issue:44230", + "right": "issue:45310", + "accept": false, + "reason": "Both mention Qwen3.5 MoE, but one is fp8 inference support and the other is a from_pretrained regression on transformers>=5.4.0. Different change requests and different failure modes." + }, + { + "left": "issue:31515", + "right": "issue:33453", + "accept": false, + "reason": "Both involve loading, but one is a performance complaint and the other is a tokenizer-loading regression. Not the same concrete bug." + }, + { + "left": "issue:43298", + "right": "issue:44230", + "accept": false, + "reason": "Continuous batching not supporting audio models is a serving capability gap; fp8 Qwen3/VL/MoE support is an inference-format/model support request. Different underlying work." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 2, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-13244-3", + "nodes": [ + "issue:13244", + "issue:15980", + "issue:16618" + ], + "soft_pairs": [ + "issue:13244|issue:16618", + "issue:15980|issue:16618" + ], + "prepared_review_unit_hash": "1118fc75400c30b705a6ee05dc49d60d79e06a9989d9cbdcc3b729f0c30480fe", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 2, + "serialized_chars": 2478, + "estimated_input_tokens": 620, + "estimated_eval_tokens": 1496 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Mostly a loose model-loading/support cluster, with one clearly unrelated tokenization issue. The two download/load complaints are superficially similar but not enough to treat as duplicates from the available evidence.", + "confidence": 0.63, + "canonical_issue_reason": "Issue #16618 is the broadest and most representative model-loading complaint in the set; it best fits the likely shared theme of failure to load a pretrained model.", + "canonical_pr_reason": null, + "best_issue_reason": "#16618 is the best global issue candidate because it is the clearest central user-facing failure in this cluster, whereas #15980 is specifically about auth/error messaging and #13244 is unrelated tokenization behavior.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:13244", + "right": "issue:16618", + "accept": false, + "reason": "Tapas tokenization mismatch is a different problem from failing to load bert-base-uncased; no evidence of the same underlying bug or code path." + }, + { + "left": "issue:15980", + "right": "issue:16618", + "accept": false, + "reason": "Both mention model download/load problems, but #15980 is specifically an unauthenticated private-model error message issue, not the same concrete failure as #16618." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 3, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-26350-2", + "nodes": [ + "issue:26350", + "issue:33260" + ], + "soft_pairs": [ + "issue:26350|issue:33260" + ], + "prepared_review_unit_hash": "429e93d0948115d40cf999ec3818d374c39600269ca89cf434fc23b883b3fe3c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1589, + "estimated_input_tokens": 398, + "estimated_eval_tokens": 1052 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are related only at a high level (architecture support requests), but they target different features: Flash Attention 2 vs GGUF. They should not be treated as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": "No canonical duplicate here: the issues ask for different capabilities and would require different implementation work.", + "canonical_pr_reason": null, + "best_issue_reason": "26350 is the more established thread by age and discussion volume, but it is not a substitute for 33260 because the feature request is different.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:26350", + "right": "issue:33260", + "accept": false, + "reason": "Different underlying requests: Flash Attention 2 support for more architectures vs GGUF support for more architectures. Similar subsystem, but not the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 4, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-27088-5", + "nodes": [ + "issue:27088", + "issue:28936", + "issue:43834", + "issue:44365", + "issue:44370" + ], + "soft_pairs": [ + "issue:27088|issue:44370", + "issue:27088|issue:44365", + "issue:27088|issue:43834", + "issue:27088|issue:28936", + "issue:28936|issue:43834", + "issue:28936|issue:44365", + "issue:28936|issue:44370", + "issue:43834|issue:44370", + "issue:43834|issue:44365", + "issue:44365|issue:44370" + ], + "prepared_review_unit_hash": "87e9a2323d8383181f033d99506b96e0f7a28838c7b3fe4ac8eafea24559c3b3", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 10, + "serialized_chars": 6442, + "estimated_input_tokens": 1611, + "estimated_eval_tokens": 3478 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 10, + "serialized_chars": 6442, + "estimated_input_tokens": 1611, + "estimated_eval_tokens": 3478 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is not a duplicate set: the visible language-specific issues are for different translations, and the near-identical placeholder issues lack enough evidence to safely merge as the same request.", + "confidence": 0.93, + "canonical_issue_reason": "No safe canonical issue: the concrete requests are for different languages (Turkish vs Spanish), and the generic placeholder issues do not provide enough body/target evidence to prove they are the same underlying task.", + "canonical_pr_reason": null, + "best_issue_reason": "No single issue is suitable as a global best duplicate target because these look like separate translation requests rather than one shared bug/change.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:27088", + "right": "issue:44370", + "accept": false, + "reason": "Different translation request; only a generic placeholder title is shared, with no evidence they target the same language or underlying task." + }, + { + "left": "issue:27088", + "right": "issue:44365", + "accept": false, + "reason": "Turkish translation request vs generic placeholder issue; not enough evidence of the same concrete request." + }, + { + "left": "issue:27088", + "right": "issue:43834", + "accept": false, + "reason": "Turkish request and generic placeholder do not clearly describe the same translation work." + }, + { + "left": "issue:27088", + "right": "issue:28936", + "accept": false, + "reason": "Different languages (Turkish vs Spanish), so these are separate translation issues." + }, + { + "left": "issue:28936", + "right": "issue:43834", + "accept": false, + "reason": "Spanish translation request is distinct from the generic placeholder; no clear same underlying task." + }, + { + "left": "issue:28936", + "right": "issue:44365", + "accept": false, + "reason": "Different translation request; the placeholder title is not enough to merge them." + }, + { + "left": "issue:28936", + "right": "issue:44370", + "accept": false, + "reason": "Spanish request and generic placeholder are not shown to be the same issue." + }, + { + "left": "issue:43834", + "right": "issue:44370", + "accept": false, + "reason": "Near-identical placeholder titles, but no body/target evidence proving the same language or same change." + }, + { + "left": "issue:43834", + "right": "issue:44365", + "accept": false, + "reason": "Generic placeholder issues may represent separate language requests; identical titles alone are insufficient to deduplicate." + }, + { + "left": "issue:44365", + "right": "issue:44370", + "accept": false, + "reason": "The issues are indistinguishable by title, but there is no concrete evidence they are the same underlying translation request." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 5, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-36817-16", + "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": "0cbd7832eec0b27ee8685d9fe7a6bb9dc9d27713b7d6db78400976c423d9475a", + "component_budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 36, + "serialized_chars": 21659, + "estimated_input_tokens": 5415, + "estimated_eval_tokens": 11086 + }, + "budget": { + "node_count": 16, + "item_count": 16, + "soft_pair_count": 36, + "serialized_chars": 21659, + "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 a collection of distinct model-request issues, mostly asking to add support for different architectures or variants. They are similar in wording but not the same underlying request, so no soft-edge duplicates are accepted.", + "confidence": 0.44, + "canonical_issue_reason": "Issue 45322 is a clear, well-scoped model request with a detailed title and body, making it a reasonable representative for this broadly similar cluster.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 45322 is the strongest single anchor here because it is specific, self-contained, and clearly formatted as a model-add request.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:36817", + "right": "issue:41211", + "accept": false, + "reason": "Different models: EuroBert vs DEIMv2. Similar request pattern, not the same change." + }, + { + "left": "issue:36817", + "right": "issue:43671", + "accept": false, + "reason": "Different models/tasks: EuroBert vs Qwen3-TTS. Not a duplicate request." + }, + { + "left": "issue:36817", + "right": "issue:36879", + "accept": false, + "reason": "Different models: EuroBert vs RF-DETR. Same broad theme only." + }, + { + "left": "issue:36879", + "right": "issue:41211", + "accept": false, + "reason": "Different models: RF-DETR vs DEIMv2." + }, + { + "left": "issue:36879", + "right": "issue:43909", + "accept": false, + "reason": "Different models: RF-DETR vs LFM2.5 Audio 1.5B." + }, + { + "left": "issue:36879", + "right": "issue:44121", + "accept": false, + "reason": "Different models: RF-DETR vs OpenAI weight-sparse transformer." + }, + { + "left": "issue:36879", + "right": "issue:43837", + "accept": false, + "reason": "Different models/tasks: RF-DETR vs Qwen3-ASR." + }, + { + "left": "issue:36879", + "right": "issue:43671", + "accept": false, + "reason": "Different models/tasks: RF-DETR vs Qwen3-TTS." + }, + { + "left": "issue:37015", + "right": "issue:43837", + "accept": false, + "reason": "Different models/tasks: NeoBERT vs Qwen3-ASR." + }, + { + "left": "issue:37015", + "right": "issue:44121", + "accept": false, + "reason": "Different models: NeoBERT vs circuit-sparsity transformer." + }, + { + "left": "issue:37015", + "right": "issue:43909", + "accept": false, + "reason": "Different models/tasks: NeoBERT vs LFM2.5 Audio 1.5B." + }, + { + "left": "issue:36879", + "right": "issue:37015", + "accept": false, + "reason": "Different models: RF-DETR vs NeoBERT." + }, + { + "left": "issue:37015", + "right": "issue:42503", + "accept": false, + "reason": "Different models: NeoBERT vs ModernVBERT." + }, + { + "left": "issue:41115", + "right": "issue:43837", + "accept": false, + "reason": "Different models/tasks: MiniCPM3 architecture vs Qwen3-ASR." + }, + { + "left": "issue:41115", + "right": "issue:44121", + "accept": false, + "reason": "Different models: MiniCPM3 vs circuit-sparsity transformer." + }, + { + "left": "issue:41115", + "right": "issue:42503", + "accept": false, + "reason": "Different models: MiniCPM3 vs ModernVBERT." + }, + { + "left": "issue:41115", + "right": "issue:45027", + "accept": false, + "reason": "Different models/tasks: MiniCPM3 vs Voxtral TTS." + }, + { + "left": "issue:41211", + "right": "issue:43671", + "accept": false, + "reason": "Different models/tasks: DEIMv2 vs Qwen3-TTS." + }, + { + "left": "issue:42503", + "right": "issue:43837", + "accept": false, + "reason": "Different models/tasks: ModernVBERT vs Qwen3-ASR." + }, + { + "left": "issue:42503", + "right": "issue:43671", + "accept": false, + "reason": "Different models/tasks: ModernVBERT vs Qwen3-TTS." + }, + { + "left": "issue:42503", + "right": "issue:42738", + "accept": false, + "reason": "Different requests: ModernVBERT models vs generic BERT-like models with RoPE." + }, + { + "left": "issue:42738", + "right": "issue:45027", + "accept": false, + "reason": "Different requests: BERT-like RoPE models vs Voxtral TTS." + }, + { + "left": "issue:43671", + "right": "issue:43837", + "accept": false, + "reason": "Related family, but different concrete support: Qwen3-TTS vs Qwen3-ASR." + }, + { + "left": "issue:43671", + "right": "issue:45027", + "accept": false, + "reason": "Different models/tasks: Qwen3-TTS vs Voxtral TTS." + }, + { + "left": "issue:43671", + "right": "issue:45177", + "accept": false, + "reason": "Different models: Qwen3-TTS vs DeepSeek-OCR2." + }, + { + "left": "issue:43671", + "right": "issue:44121", + "accept": false, + "reason": "Different models/tasks: Qwen3-TTS vs circuit-sparsity transformer." + }, + { + "left": "issue:43671", + "right": "issue:43909", + "accept": false, + "reason": "Different models/tasks: Qwen3-TTS vs LFM2.5 Audio 1.5B." + }, + { + "left": "issue:43837", + "right": "issue:43909", + "accept": false, + "reason": "Different models/tasks: Qwen3-ASR vs LFM2.5 Audio 1.5B." + }, + { + "left": "issue:43837", + "right": "issue:45322", + "accept": false, + "reason": "Different models/tasks: Qwen3-ASR vs EUPE." + }, + { + "left": "issue:43837", + "right": "issue:44121", + "accept": false, + "reason": "Different models/tasks: Qwen3-ASR vs circuit-sparsity transformer." + }, + { + "left": "issue:43909", + "right": "issue:44205", + "accept": false, + "reason": "Different models/tasks: LFM2.5 Audio 1.5B vs SAM3-LiteText." + }, + { + "left": "issue:44121", + "right": "issue:45322", + "accept": false, + "reason": "Different models: circuit-sparsity transformer vs EUPE." + }, + { + "left": "issue:43837", + "right": "issue:45027", + "accept": false, + "reason": "Different models/tasks: Qwen3-ASR vs Voxtral TTS." + }, + { + "left": "issue:37015", + "right": "issue:45027", + "accept": false, + "reason": "Different models/tasks: NeoBERT vs Voxtral TTS." + }, + { + "left": "issue:43837", + "right": "issue:45177", + "accept": false, + "reason": "Different models/tasks: Qwen3-ASR vs DeepSeek-OCR2." + }, + { + "left": "issue:45322", + "right": "issue:46133", + "accept": false, + "reason": "Different model requests: EUPE vs TIPSv2. Same general domain, not the same change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 6, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-37934-10", + "nodes": [ + "issue:37934", + "issue:42936", + "issue:43240", + "issue:43668", + "issue:43717", + "issue:43813", + "issue:44031", + "issue:46018", + "issue:46143", + "issue:46144" + ], + "soft_pairs": [ + "issue:37934|issue:43668", + "issue:42936|issue:43240", + "issue:42936|issue:46018", + "issue:42936|issue:46144", + "issue:42936|issue:44031", + "issue:42936|issue:43813", + "issue:43240|issue:46144", + "issue:43240|issue:43668", + "issue:43668|issue:44031", + "issue:43668|issue:43813", + "issue:42936|issue:43668", + "issue:43717|issue:46144", + "issue:46143|issue:46144" + ], + "prepared_review_unit_hash": "86c9244c776f418640b0841225da650df5d637bd14315c4f716f948f91ade438", + "component_budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 13, + "serialized_chars": 10295, + "estimated_input_tokens": 2574, + "estimated_eval_tokens": 5404 + }, + "budget": { + "node_count": 10, + "item_count": 10, + "soft_pair_count": 13, + "serialized_chars": 10295, + "estimated_input_tokens": 2574, + "estimated_eval_tokens": 5404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster appears to be a set of unrelated model/config/documentation bugs that only share broad library vocabulary; none of the soft pairs look like true duplicates or mergeable into one fix.", + "confidence": 0.93, + "canonical_issue_reason": "If a single issue must represent the set, issue 46144 is the most concrete and best-supported report, with the strongest discussion/inbound activity and the clearest actionable code-path bug; the others are separate problems rather than duplicates.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 46144 is the strongest standalone candidate: it describes a specific RoFormer attention-interface defect, has some external attention, and is more actionable than the other reports in this group.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:37934", + "right": "issue:43668", + "accept": false, + "reason": "Different model components and different bug types: Llama4 norm-vs-RMS question vs a ModernBERT config type-hint issue." + }, + { + "left": "issue:42936", + "right": "issue:43240", + "accept": false, + "reason": "Mask2former ignore_value usage and loss-function kwargs forwarding are separate code paths and separate fixes." + }, + { + "left": "issue:42936", + "right": "issue:46018", + "accept": false, + "reason": "Unrelated model behaviors: Mask2former loss handling vs DeepSeek-V4 shared expert gating." + }, + { + "left": "issue:42936", + "right": "issue:46144", + "accept": false, + "reason": "No shared concrete bug; Mask2former ignore_value and RoFormer attention-interface behavior are unrelated." + }, + { + "left": "issue:42936", + "right": "issue:44031", + "accept": false, + "reason": "Model loss-parameter issue vs tokenizer regex warning; different subsystems and fixes." + }, + { + "left": "issue:42936", + "right": "issue:43813", + "accept": false, + "reason": "Different concerns entirely: Mask2former loss usage vs a quantization-string typo." + }, + { + "left": "issue:43240", + "right": "issue:46144", + "accept": false, + "reason": "Both mention kwargs, but one is about loss-function forwarding and the other about RoFormer methods/attention wiring; not the same bug." + }, + { + "left": "issue:43240", + "right": "issue:43668", + "accept": false, + "reason": "Loss kwargs propagation and a config type-hint typo are unrelated issues." + }, + { + "left": "issue:43668", + "right": "issue:44031", + "accept": false, + "reason": "Type-hint correctness and tokenizer regex warnings are unrelated." + }, + { + "left": "issue:43668", + "right": "issue:43813", + "accept": false, + "reason": "Both are small correctness issues, but one is a config typing mistake and the other is a typo in a different code path." + }, + { + "left": "issue:42936", + "right": "issue:43668", + "accept": false, + "reason": "Mask2former ignore_value behavior and ModernBERT config typing are not the same underlying bug." + }, + { + "left": "issue:43717", + "right": "issue:46144", + "accept": false, + "reason": "Mamba init_weights behavior and RoFormer attention-interface behavior are distinct model-specific bugs." + }, + { + "left": "issue:46143", + "right": "issue:46144", + "accept": false, + "reason": "Closely related RoFormer reports, but one is about kwargs not being passed through methods and the other about attention not using the interface; related area, different concrete fix." + } + ] + }, + "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": "767c8dacf4bd1f2c5b26e9ebffa3981cd7b356e7c6f4efa9949f496c13e7c2ef", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1479, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1479, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues are not duplicates: one requests Set Block Decoding, while the other asks for DeepSeek v3.2 support. They describe different features/problems with no shared targets or explicit linkage.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:41084", + "right": "issue:43037", + "accept": false, + "reason": "Different underlying requests: decoding behavior vs model support. No shared code path, filenames, or explicit relation indicating the same bug/change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "3fa11e03f42da98482734e5ed7dce4d6f74543d23229ac0765455c8633ffc54b", + "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 look like two separate Trainer enhancement issues: one about making `remove_unused_column` more customizable, and the other about correctly estimating FLOPS for custom trainers. They share the Trainer/custom-trainer theme but not the same bug or change.", + "confidence": 0.94, + "canonical_issue_reason": "Issue #43598 is the cleaner canonical issue for its own topic because it is a focused Trainer customization request with a clear, specific hook (`remove_unused_column`).", + "canonical_pr_reason": null, + "best_issue_reason": "Neither issue is a duplicate of the other, so there is no global best issue for the cluster; if forced to pick one representative, #43598 is more concrete and narrowly scoped.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43598", + "right": "issue:43600", + "accept": false, + "reason": "Both involve custom trainers in Trainer, but they address different underlying problems: one is about unused-column removal customization, the other about FLOPS estimation. They are not the same bug/change and should not be merged as duplicates." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "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": "66e707e517b28e39997d8cd378a0b3b3a062cf4f755ca00b34bfa2183808a516", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1612, + "estimated_input_tokens": 403, + "estimated_eval_tokens": 1062 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are not duplicates: one requests Qwen3-Omni registration in generic auto-model classes, while the other asks for sequence-classification support for Qwen3.5. They involve different model families and different API surfaces/tasks.", + "confidence": 0.96, + "canonical_issue_reason": "Issue 43684 is the broader, more central registry gap request and is the better representative of a model-registration problem.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue 43684 is the clearest standalone issue to keep as the canonical issue because it describes a general AutoModel/AutoModelForConditionalGeneration registration addition.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43684", + "right": "issue:44405", + "accept": false, + "reason": "Different underlying change request: Qwen3-Omni auto-model registration vs Qwen3.5 sequence-classification support. Different model family and different code path." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "32c7832b1cbeb78151b3a483f4355d28cf5b7cc3d81c6bec504b425ff90fcc3a", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4259, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 4259, + "estimated_input_tokens": 1065, + "estimated_eval_tokens": 2386 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The only strong duplicate signal is the near-simultaneous pair 43845/43846, both titled \"huggingface \". 43859 shares the same title but is separated by many hours, so it is a weaker match and should not be merged on title alone. 43835 (\"Fett\") is unrelated.", + "confidence": 0.76, + "canonical_issue_reason": "Issue 43845 is the best canonical representative for the clear duplicate pair because it is the earlier of the two near-identical \"huggingface \" reports.", + "canonical_pr_reason": null, + "best_issue_reason": "43845 is the clearest representative of the duplicate set: it matches 43846 closely and is the earliest of that tight duplicate pair.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:43835", + "right": "issue:43845", + "accept": false, + "reason": "\"Fett\" is unrelated to the repeated \"huggingface \" reports; no evidence they describe the same issue." + }, + { + "left": "issue:43835", + "right": "issue:43846", + "accept": false, + "reason": "\"Fett\" is unrelated to the repeated \"huggingface \" reports; no evidence they describe the same issue." + }, + { + "left": "issue:43835", + "right": "issue:43859", + "accept": false, + "reason": "\"Fett\" is unrelated to the repeated \"huggingface \" reports; no evidence they describe the same issue." + }, + { + "left": "issue:43845", + "right": "issue:43846", + "accept": true, + "reason": "Same title and created within minutes, which strongly suggests the same underlying duplicate issue." + }, + { + "left": "issue:43845", + "right": "issue:43859", + "accept": false, + "reason": "Same title, but the reports are separated by many hours and there is no additional evidence beyond soft similarity to confirm the same underlying issue." + }, + { + "left": "issue:43846", + "right": "issue:43859", + "accept": false, + "reason": "Same title, but the reports are separated by many hours and there is no additional evidence beyond soft similarity to confirm the same underlying issue." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The grouping is mostly grounded, but the soft-edge reasoning is not fully conservative. In particular, the 43845/43859 and 43846/43859 pairs are separated by many hours, so timing does not really support a duplicate inference there; the evidence is only the identical title/soft similarity. The summary and canonical choice for 43845 are reasonable, but the duplicate verdicts should be phrased more cautiously." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "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": "310c6b85729a16148216edff379f05ec32b7082f4305b8de3f59c4372f771e11", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 9, + "serialized_chars": 5791, + "estimated_input_tokens": 1448, + "estimated_eval_tokens": 3152 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is dominated by spammy or content-free issues with only one mildly descriptive title, and there is no evidence that any pair is the same underlying bug or request. Treat them as separate low-signal issues rather than duplicates.", + "confidence": 0.97, + "canonical_issue_reason": "Issue #44418 is the least placeholder-like item because it has a descriptive title, but it still lacks any concrete problem statement and does not match the others as a clear duplicate.", + "canonical_pr_reason": null, + "best_issue_reason": "#44418 is the best representative issue only because its title is more specific than the pure spam entries; it is not a strong canonical duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:44371", + "right": "issue:44572", + "accept": false, + "reason": "Both are just placeholder spam titles with no substantive body or issue details; no concrete shared bug or request is visible." + }, + { + "left": "issue:44371", + "right": "issue:44573", + "accept": false, + "reason": "These appear to be separate spam entries rather than the same issue, with no identifying context tying them together." + }, + { + "left": "issue:44371", + "right": "issue:44418", + "accept": false, + "reason": "One is placeholder spam and the other is a generic documentation suggestion; too little evidence that they describe the same underlying request." + }, + { + "left": "issue:44371", + "right": "issue:44937", + "accept": false, + "reason": "'Check out \"Google LLC\"' does not align with the spam placeholder title as the same underlying issue." + }, + { + "left": "issue:44418", + "right": "issue:44572", + "accept": false, + "reason": "A generic documentation enhancement and a spam placeholder are not enough to conclude a duplicate relationship." + }, + { + "left": "issue:44418", + "right": "issue:44573", + "accept": false, + "reason": "No concrete overlap in problem statement or requested change; these look like unrelated low-quality issues." + }, + { + "left": "issue:44572", + "right": "issue:44573", + "accept": false, + "reason": "Although both are placeholder titles, that alone is not enough to merge them as the same underlying issue." + }, + { + "left": "issue:44572", + "right": "issue:44937", + "accept": false, + "reason": "The titles suggest different spam/content themes, not a single shared bug or request." + }, + { + "left": "issue:44573", + "right": "issue:44937", + "accept": false, + "reason": "No shared concrete problem description or target is present; these should not be deduplicated." + } + ] + }, + "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": "00ab2e5e779fc12cf48acc5fc2b3c6c483431e5917284572e8ab1f70f8ec80e7", + "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 highly similar and likely related duplicate reports, but the packet only establishes title-level similarity and does not prove they are the same underlying proposal. Issue #45145 is the better anchor because it is earlier and already has an inbound reference.", + "confidence": 0.82, + "canonical_issue_reason": "Issue #45145 is the better canonical issue because it was created first and already has an inbound reference, so it is the likelier discussion anchor for a probable duplicate pair.", + "canonical_pr_reason": null, + "best_issue_reason": "Issue #45145 is the best global issue candidate for the same reason: it is the earliest and most referenced of the two near-identical reports.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45145", + "right": "issue:45151", + "accept": false, + "reason": "The titles are identical and the reports look very similar, but the evidence is not strong enough to confirm they are the same underlying issue. There are also small distinguishing signals, including inbound reference and activity/body-length differences." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The duplicate call is directionally plausible, but the summary overstates what the packet proves. The packet supports strong title similarity and a likely duplicate relationship, yet it does not establish that the two issues are the same underlying proposal, and it is not true that there is \u201cno distinguishing metadata beyond creation time\u201d because issue #45145 has an inbound reference while #45151 does not, and their discussion activity/body length also differ. Please soften the language to match the evidence." + }, + "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": "feab38ee0466aeb63912ce8f83c339fabd9ce085cb7cbb9572a02af7fe9ea9ce", + "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 related to 4-bit inference, but they describe different failures with different triggers and code paths, so they should not be merged as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45672", + "right": "issue:45674", + "accept": false, + "reason": "Different bugs: one is a Gemma4/audio-module torch.finfo() TypeError on uint8 weights during NF4 inference, the other is an AssertionError caused by llm_int8_skip_modules clearing the default lm_head exclusion. Same broad area, but not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 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": "c1a394bda8158756b4a565994cc44969a493062fa34978f9995c3b4d81ae4212", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1477, + "estimated_input_tokens": 370, + "estimated_eval_tokens": 996 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two closed issues with unrelated person-name/date titles and no body or linking evidence, so they do not look like duplicates or the same underlying report.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45762", + "right": "issue:45763", + "accept": false, + "reason": "Different named individuals and dates, with no shared description, references, or code context; not the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 16, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45764-2", + "nodes": [ + "issue:45764", + "issue:45765" + ], + "soft_pairs": [ + "issue:45764|issue:45765" + ], + "prepared_review_unit_hash": "7d379b7281f27058ab2bf9bdadb60710273c675e5fa391a1d28e7f874d03b264", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 1481, + "estimated_input_tokens": 371, + "estimated_eval_tokens": 998 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The two issues only share a loose title pattern (a person\u2019s name plus date) and provide no body, technical context, or linkage showing they refer to the same underlying bug/task. They should not be clustered as duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": "No strong canonical issue candidate: both are sparse, non-technical entries with different names and dates, so neither can be confidently treated as the duplicate target.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45764", + "right": "issue:45765", + "accept": false, + "reason": "Reject: the titles refer to different people and different dates, with no body or explicit linkage indicating the same underlying issue." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "issue", + "component_index": 17, + "component_count": 17, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45869-5", + "nodes": [ + "issue:45869", + "issue:45870", + "issue:45871", + "issue:45872", + "issue:45873" + ], + "soft_pairs": [ + "issue:45869|issue:45870", + "issue:45869|issue:45871", + "issue:45870|issue:45871", + "issue:45870|issue:45872", + "issue:45870|issue:45873", + "issue:45871|issue:45872", + "issue:45871|issue:45873", + "issue:45872|issue:45873" + ], + "prepared_review_unit_hash": "55d255125c68e555ba3adc3ca72b51c2279bbf06a28ea1aee3af376143d1559e", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 8, + "serialized_chars": 5726, + "estimated_input_tokens": 1432, + "estimated_eval_tokens": 3120 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two issue pairs look highly similar and are likely duplicates: 45870/45871 and 45872/45873. The packet only shows titles and body lengths, so this should be treated as strong duplicate evidence rather than a confirmed exact match. Issue 45869 is a separate topic.", + "confidence": 0.91, + "canonical_issue_reason": "45870 is the earliest item in the duplicated \"Example: Financial time series prediction with DataForge\" pair and is the cleanest representative of that issue thread; 45869 is unrelated, and 45872/45873 form a separate duplicate pair.", + "canonical_pr_reason": null, + "best_issue_reason": "45870 is the best overall issue representative because it is the earliest clear duplicate in the packet and has the most straightforward canonical title among the closely matching items.", + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "issue:45869", + "right": "issue:45870", + "accept": false, + "reason": "Different topics: a dataset bundle listing versus a financial time series prediction example." + }, + { + "left": "issue:45869", + "right": "issue:45871", + "accept": false, + "reason": "Different topics: a dataset bundle listing versus a financial time series prediction example." + }, + { + "left": "issue:45870", + "right": "issue:45871", + "accept": true, + "reason": "Same title and matching body length; very likely the same issue content, though the full bodies are not shown." + }, + { + "left": "issue:45870", + "right": "issue:45872", + "accept": false, + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." + }, + { + "left": "issue:45870", + "right": "issue:45873", + "accept": false, + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." + }, + { + "left": "issue:45871", + "right": "issue:45872", + "accept": false, + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." + }, + { + "left": "issue:45871", + "right": "issue:45873", + "accept": false, + "reason": "Different underlying topics: prediction example versus a scraping-service/DataForge request." + }, + { + "left": "issue:45872", + "right": "issue:45873", + "accept": true, + "reason": "Same title and matching body length; very likely the same issue content, though the full bodies are not shown." + } + ] + }, + "evaluator_result": { + "accept": false, + "feedback": "The summary is a bit too strong for the packet. The evidence supports that 45870/45871 and 45872/45873 are highly similar and likely duplicates (identical titles and matching body lengths), but the packet does not show the full bodies, so calling them \u201cexact duplicates\u201d overstates certainty. The statement that 45869 is a separate topic is reasonable, but the duplicate verdicts should be phrased more conservatively." + }, + "evaluator_used": true, + "retried": true, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 1, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-39895-14", + "nodes": [ + "pull_request:39895", + "pull_request:41855", + "pull_request:42668", + "pull_request:43251", + "pull_request:43254", + "pull_request:43424", + "pull_request:43823", + "pull_request:43995", + "pull_request:44827", + "pull_request:45170", + "pull_request:45213", + "pull_request:45655", + "pull_request:46048", + "pull_request:46067" + ], + "soft_pairs": [ + "pull_request:39895|pull_request:43424", + "pull_request:41855|pull_request:43823", + "pull_request:41855|pull_request:43424", + "pull_request:42668|pull_request:43823", + "pull_request:43251|pull_request:46067", + "pull_request:43251|pull_request:43823", + "pull_request:43251|pull_request:44827", + "pull_request:43424|pull_request:46067", + "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:43823|pull_request:46067", + "pull_request:43995|pull_request:44827", + "pull_request:43995|pull_request:45213", + "pull_request:44827|pull_request:46067", + "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": "f3e9fdc8479def9b7ae00e8aebac33c70ce931040c8143465ce83af123d1afd6", + "component_budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 22, + "serialized_chars": 38366, + "estimated_input_tokens": 9592, + "estimated_eval_tokens": 19440 + }, + "budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 22, + "serialized_chars": 38366, + "estimated_input_tokens": 9592, + "estimated_eval_tokens": 19440 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "The cluster is mostly unrelated PRs spanning model additions, refactors, and test/docs changes. The only true duplicate pair is the two fixed_cross_entropy PRs, which patch the same loss helper with the same kwargs support.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "pull_request:43251 is the best representative for the loss fix: it has the explicit issue link, targets the exact bug in fixed_cross_entropy, and its change is the cleanest description of the shared code-path fix.", + "best_issue_reason": null, + "best_pr_reason": "pull_request:43251 is the strongest overall PR in this cluster for duplicate triage because it most directly and clearly addresses the shared bug; the other near-match (43254) is just an alternate wording of the same patch.", + "soft_edge_verdicts": [ + { + "left": "pull_request:39895", + "right": "pull_request:43424", + "accept": false, + "reason": "Unrelated PRs: VideoPrism model/docs addition vs Executorch dynamic-shape test." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43823", + "accept": false, + "reason": "Different models and different changes: Mistral tokenizer methods vs adding MobileLLM." + }, + { + "left": "pull_request:41855", + "right": "pull_request:43424", + "accept": false, + "reason": "Tokenizer API fix vs Executorch export test; no shared bug or code path." + }, + { + "left": "pull_request:42668", + "right": "pull_request:43823", + "accept": false, + "reason": "Processor loading robustness changes for existing models vs adding a new model implementation." + }, + { + "left": "pull_request:43251", + "right": "pull_request:46067", + "accept": false, + "reason": "Loss helper kwargs fix vs conversion/loading prefix handling; different subsystems and bugs." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43823", + "accept": false, + "reason": "Cross-entropy helper fix vs MobileLLM addition; unrelated." + }, + { + "left": "pull_request:43251", + "right": "pull_request:44827", + "accept": false, + "reason": "Loss helper kwargs fix vs Mistral4 test/model adjustments; different issues." + }, + { + "left": "pull_request:43424", + "right": "pull_request:46067", + "accept": false, + "reason": "Executorch dynamic-shape test vs base_model_prefix conversion fix; unrelated." + }, + { + "left": "pull_request:43424", + "right": "pull_request:44827", + "accept": false, + "reason": "Executorch export testing vs Mistral4 test/model changes; not the same bug." + }, + { + "left": "pull_request:43424", + "right": "pull_request:43823", + "accept": false, + "reason": "Executorch export test vs new MobileLLM model; unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45655", + "accept": false, + "reason": "New model implementation vs config resolution order fix; no shared concrete change." + }, + { + "left": "pull_request:43823", + "right": "pull_request:43995", + "accept": false, + "reason": "New model addition vs Falcon output-interface refactor; different targets." + }, + { + "left": "pull_request:43823", + "right": "pull_request:46067", + "accept": false, + "reason": "New model addition vs conversion prefix-loading fix; unrelated." + }, + { + "left": "pull_request:43995", + "right": "pull_request:44827", + "accept": false, + "reason": "Falcon output-capture refactor vs Mistral4 fixes; different code paths." + }, + { + "left": "pull_request:43995", + "right": "pull_request:45213", + "accept": false, + "reason": "Falcon refactor vs model-creation skill/meta PR; not a duplicate bug fix." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46067", + "accept": false, + "reason": "Mistral4 test/model changes vs conversion base_model_prefix loading fix; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45170", + "accept": false, + "reason": "Mistral4 fixes vs layrnorm/layernorm rename cleanup; different underlying issue." + }, + { + "left": "pull_request:44827", + "right": "pull_request:46048", + "accept": false, + "reason": "Mistral4 fixes vs TDT loss kernel addition; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45655", + "accept": false, + "reason": "Mistral4 fixes vs config resolution order fix; not the same change." + }, + { + "left": "pull_request:44827", + "right": "pull_request:45213", + "accept": false, + "reason": "Mistral4-related code vs model-creation skill/meta PR; unrelated." + }, + { + "left": "pull_request:44827", + "right": "pull_request:43254", + "accept": false, + "reason": "Mistral4 fixes vs cross_entropy kwargs patch; unrelated." + }, + { + "left": "pull_request:43823", + "right": "pull_request:45213", + "accept": false, + "reason": "New MobileLLM model vs meta skill PR; not a duplicate." + }, + { + "left": "pull_request:43251", + "right": "pull_request:43254", + "accept": true, + "reason": "Both PRs fix the same fixed_cross_entropy helper to forward cross_entropy kwargs on the same code path, and could be merged into a single change." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet, and the soft-edge verdicts stay conservative. The only accepted pair, pull_request:43251 with pull_request:43254, is clearly mergeable into one PR for the same fixed_cross_entropy kwargs bugfix. The rest are reasonable non-duplicate rejections based on distinct subsystems, targets, or change types." + }, + "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": "b7ebb48bdff3221cfad0c0410f23a71d3907c42ae57a85242833cf66cef027bf", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7224, + "estimated_input_tokens": 1806, + "estimated_eval_tokens": 3868 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 7224, + "estimated_input_tokens": 1806, + "estimated_eval_tokens": 3868 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related VibeVoice pull requests, but they are not the same change: one implements the base VibeVoice integration, while the other adds a separate realtime variant plus acoustic tokenizer work. The overlap is mostly shared model scaffolding, not a single duplicate fix or mergeable one-PR patch.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #40546 is the main baseline VibeVoice implementation and has substantial review activity; #42781 is a draft realtime extension with additional tokenizer/generation changes, so it is not the same concrete change.", + "best_issue_reason": null, + "best_pr_reason": "PR #40546 is the best representative PR for the cluster because it is the primary full VibeVoice integration and is more mature than the draft realtime follow-up.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40546", + "right": "pull_request:42781", + "accept": false, + "reason": "Shared filenames show common VibeVoice groundwork, but the titles and diff scope differ: base VibeVoice implementation versus realtime-specific additions with acoustic tokenizer support. They are related, not the same underlying change, and would not plausibly be merged into one PR without conflating distinct features." + } + ] + }, + "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": "8c869560bef32a5fbd2d7ceb4a4fa3968c80e790cd5b73dbc241fb0f48f25384", + "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 unrelated pull requests: one adds TimesFM covariate forecasting, the other adds DINOv3 image-classification support and docs. They do not fix the same bug or implement the same change.", + "confidence": 0.99, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "No single best PR for duplication triage; the PRs target different models, code paths, and user-facing features.", + "soft_edge_verdicts": [ + { + "left": "pull_request:40755", + "right": "pull_request:41224", + "accept": false, + "reason": "Different subsystems and different changes: TimesFM covariate forecasting vs DINOv3 image classification support. They are not the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "9a5320ea4eb7befdef29cb035d4fd22dd6098546d88639bfc1cff8c3d8fae09f", + "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-addition PRs with overlapping scaffolding but different target models and code paths. The soft similarities are only at the shared Transformers integration pattern level, not the same underlying change.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No canonical PR: the three PRs add different models (DeepSeek OCR, Isaac, Kimi2-6) and are not mergeable as one duplicate fix/change.", + "best_issue_reason": null, + "best_pr_reason": "#45630 is the strongest representative of the cluster only in the sense that it is the most complete integration PR, but it is still distinct from the others and should not be treated as a duplicate.", + "soft_edge_verdicts": [ + { + "left": "pull_request:41797", + "right": "pull_request:45186", + "accept": false, + "reason": "Different model additions for DeepSeek OCR vs Isaac; shared auto-registration/docs patterns do not indicate the same bug or change." + }, + { + "left": "pull_request:45186", + "right": "pull_request:45630", + "accept": false, + "reason": "Different model additions for Isaac vs Kimi2-6; they share framework plumbing but fix/add separate model integrations 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": 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": "02ef4b642c6eac7fb0ebdcfc1bf6becbdb2df91d4418a742897eccdc0919ca44", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5935, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5935, + "estimated_input_tokens": 1484, + "estimated_eval_tokens": 3224 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Two open PRs both update GitHub Actions workflow dependencies, but they target different upgrades: one is an umbrella Node 24 compatibility sweep, the other specifically bumps docker/build-push-action to v6. They overlap in some workflow files but do not fix the same concrete change.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 42975 is the better canonical representative because it has the broader, explicit goal of making workflows compatible with Node 24 and updates multiple core GitHub Actions references across many workflows.", + "best_issue_reason": null, + "best_pr_reason": "PR 42975 is the best overall representative for the cluster since it captures the broader maintenance effort; PR 42976 is a narrower action-version bump focused on docker/build-push-action.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42975", + "right": "pull_request:42976", + "accept": false, + "reason": "They are not the same underlying change: 42975 upgrades several GitHub Actions for Node 24 compatibility, while 42976 only bumps docker/build-push-action versions. Shared workflow files alone are not enough to merge these 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": 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-42978-4", + "nodes": [ + "pull_request:42978", + "pull_request:43448", + "pull_request:43451", + "pull_request:45490" + ], + "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" + ], + "prepared_review_unit_hash": "c9f5e2d5f2aa3c90fcded3f17e12d7f10ef8c924d3878a11e2a4d8750da72cdd", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 14178, + "estimated_input_tokens": 3545, + "estimated_eval_tokens": 7346 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 6, + "serialized_chars": 14178, + "estimated_input_tokens": 3545, + "estimated_eval_tokens": 7346 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs all add different model integrations (ViT NEPA, Molmo, Molmo2, CTSM) and only share broad scaffolding like docs/auto-registry updates, so they are not duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 43451 (Molmo2) is the most substantial and active of the set, so it is the best anchor if a single representative PR is needed, but it is not the same change as the others.", + "best_issue_reason": null, + "best_pr_reason": "PR 43451 is the strongest standalone PR in the cluster: largest review activity and a full model stack addition, making it the best representative despite the cluster containing unrelated model additions.", + "soft_edge_verdicts": [ + { + "left": "pull_request:42978", + "right": "pull_request:43448", + "accept": false, + "reason": "Different model additions (ViT NEPA vs Molmo) with distinct model files and tests; shared registry/docs edits are incidental." + }, + { + "left": "pull_request:42978", + "right": "pull_request:43451", + "accept": false, + "reason": "ViT NEPA and Molmo2 are separate architectures and code paths; overlap is only in generic auto/docs plumbing." + }, + { + "left": "pull_request:42978", + "right": "pull_request:45490", + "accept": false, + "reason": "ViT NEPA and CTSM are unrelated model integrations, not one underlying bug or change." + }, + { + "left": "pull_request:43448", + "right": "pull_request:43451", + "accept": false, + "reason": "Molmo and Molmo2 are different models with different implementations; they cannot plausibly be merged as one fix/change." + }, + { + "left": "pull_request:43448", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo and CTSM touch separate model families and code paths; common files are only shared integration points." + }, + { + "left": "pull_request:43451", + "right": "pull_request:45490", + "accept": false, + "reason": "Molmo2 and CTSM are distinct model additions with different affected modules and tests, so this is not a duplicate pair." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 8, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43291-4", + "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": "8ca51f6891c4288b5e18dbfb41ab24e597ff2308c60248069ba629d146fbacf9", + "component_budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 9391, + "estimated_input_tokens": 2348, + "estimated_eval_tokens": 4952 + }, + "budget": { + "node_count": 4, + "item_count": 4, + "soft_pair_count": 4, + "serialized_chars": 9391, + "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: they touch different subsystems and fix unrelated problems. The only clearly substantive PR among them is the FP8 dequantization fix.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical duplicate: the PRs are unrelated (Whisper decoding/tests, CLIP/ViT test stubs, FP8 dequantization, and FSDP tied-weights cleanup/fix).", + "best_issue_reason": null, + "best_pr_reason": "PR 44815 is the strongest representative because it contains a concrete multi-file code fix in model loading and FP8 quantization paths, with supporting tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43291", + "right": "pull_request:43665", + "accept": false, + "reason": "Different areas and different changes: Whisper tokenizer logic vs unrelated CLIP/ViT test edits." + }, + { + "left": "pull_request:43291", + "right": "pull_request:44815", + "accept": false, + "reason": "Whisper decoding/timestamp handling is unrelated to FP8 dequantization and model-loading changes." + }, + { + "left": "pull_request:43665", + "right": "pull_request:46102", + "accept": false, + "reason": "Placeholder test failures in CLIP/ViT do not match an FSDP tied-weights fix." + }, + { + "left": "pull_request:43665", + "right": "pull_request:44815", + "accept": false, + "reason": "Generic test changes are not the same underlying bug as the FP8 dequantization/model-loading fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "81087537cf084555e9063b262cb4072fe42a4975497e8f528d533716c384b6d5", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 11339, + "estimated_input_tokens": 2835, + "estimated_eval_tokens": 5926 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 11339, + "estimated_input_tokens": 2835, + "estimated_eval_tokens": 5926 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "One clear PR duplicate pair exists: 46100 and 46114 are the same ViT test-migration edit. The other PRs are lookalikes only and differ in scope/content, so they should not be merged as duplicates.", + "confidence": 0.96, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 46100 is the cleanest representative of the duplicated ViT test-migration change: it matches 46114 line-for-line in tests/models/vit/test_modeling_vit.py, while the rest of the cluster is unrelated.", + "best_issue_reason": null, + "best_pr_reason": "PR 46100 is the best canonical PR because it is the earlier, straightforward version of the duplicated patch and its diff is identical to 46114.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43488", + "right": "pull_request:43532", + "accept": false, + "reason": "Same broad repo area and one shared test filename, but the changes are unrelated: 43488 is a setup/dependency/version tweak, while 43532 changes workflow/test infrastructure across several files." + }, + { + "left": "pull_request:43488", + "right": "pull_request:46114", + "accept": false, + "reason": "Different underlying changes. 43488 is a repo-bot/setup PR with dependency/version edits; 46114 is a tiny ViT test-migration change in tests/models/vit/test_modeling_vit.py." + }, + { + "left": "pull_request:43532", + "right": "pull_request:46114", + "accept": false, + "reason": "They both touch test-related files, but the concrete change is different: 43532 modifies workflows and test utilities, while 46114 only adds failing ViT test stubs." + }, + { + "left": "pull_request:45932", + "right": "pull_request:46100", + "accept": false, + "reason": "Not the same bug/change. 45932 is a broad multi-model/config/workflow PR, whereas 46100 only edits a single ViT test file with dummy assertions." + }, + { + "left": "pull_request:46100", + "right": "pull_request:46114", + "accept": true, + "reason": "These are the same underlying change: identical edits to tests/models/vit/test_modeling_vit.py adding the same test-migration assertions, so they could plausibly be merged into one PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The cluster summary is grounded in the packet. PRs 46100 and 46114 have identical diffs in tests/models/vit/test_modeling_vit.py, so treating them as the duplicate pair is well supported. The other pairs are rejected conservatively based on differing scope and content, not just filename overlap." + }, + "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": "ec6d1d9aa72adba06ef7bc037729bb4f9ba9e9783de8cb9341faf5cd2f9360a6", + "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 share pipeline/documentation scaffolding but implement different features: promptable concept segmentation vs promptable visual segmentation. They are related, not duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": null, + "soft_edge_verdicts": [ + { + "left": "pull_request:43612", + "right": "pull_request:43613", + "accept": false, + "reason": "Both add similar segmentation-pipeline plumbing, but they target different underlying changes and code paths: concept segmentation vs visual segmentation, with different model integrations and test files. They should stay separate." + } + ] + }, + "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": "47e97af7371d620064507c8470b80e5ab10877bd2efa10299f797e3dfe376c23", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4293, + "estimated_input_tokens": 1074, + "estimated_eval_tokens": 2404 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4293, + "estimated_input_tokens": 1074, + "estimated_eval_tokens": 2404 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs modify `src/transformers/trainer.py` and add override hooks, but they target unrelated trainer behaviors: gradient-accumulation loss scaling vs. signature-based column filtering.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: these are separate trainer extension-point changes, not duplicate fixes for the same code path.", + "best_issue_reason": null, + "best_pr_reason": "PR 43651 is the more self-contained change, but it is still distinct from 43663 and does not subsume it.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43651", + "right": "pull_request:43663", + "accept": false, + "reason": "Reject: same file and similar helper-style refactor, but the underlying behavior is different. 43651 changes loss scaling during training steps; 43663 changes which input columns are kept by inspecting the model signature." + } + ] + }, + "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": "127355f5fb6fe5ec7cef7191486be3a58e65d6bbe3849c33276fec28a4aa93d0", + "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": "Both PRs touch `src/transformers/cli/serve.py` and mention a TypeAdapter-related NameError, but they appear to fix different annotation/import problems rather than the same concrete change.", + "confidence": 0.8, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "PR #43842 is the more direct, narrowly scoped fix for the explicit `TypeAdapter` NameError when `pydantic` is absent; PR #43656 is a broader typing/annotation cleanup and is not the same patch.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43656", + "right": "pull_request:43842", + "accept": false, + "reason": "Same file and related symptom, but different fixes: #43656 is a broader annotations/__future__ typing change, while #43842 specifically gates `TypeAdapter` behind `TYPE_CHECKING` and quotes the annotation. They are not the same concrete code-path 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": 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:44024", + "pull_request:44044", + "pull_request:44056", + "pull_request:44059", + "pull_request:44066", + "pull_request:44068", + "pull_request:44071", + "pull_request:44072", + "pull_request:44073", + "pull_request:44074", + "pull_request:44076", + "pull_request:44085", + "pull_request:44086", + "pull_request:44098", + "pull_request:44101", + "pull_request:44116", + "pull_request:44129", + "pull_request:44154", + "pull_request:44161", + "pull_request:44722" + ], + "soft_pairs": [ + "pull_request:43996|pull_request:44085", + "pull_request:43996|pull_request:44044", + "pull_request:44066|pull_request:44085", + "pull_request:44007|pull_request:44072", + "pull_request:44072|pull_request:44722", + "pull_request:44066|pull_request:44072", + "pull_request:44013|pull_request:44044", + "pull_request:44066|pull_request:44086", + "pull_request:44066|pull_request:44071", + "pull_request:44018|pull_request:44068", + "pull_request:44066|pull_request:44068", + "pull_request:44018|pull_request:44066", + "pull_request:44019|pull_request:44722", + "pull_request:44086|pull_request:44722", + "pull_request:44019|pull_request:44085", + "pull_request:44018|pull_request:44019", + "pull_request:44019|pull_request:44071", + "pull_request:44068|pull_request:44722", + "pull_request:44068|pull_request:44116", + "pull_request:44071|pull_request:44722", + "pull_request:44018|pull_request:44116", + "pull_request:44018|pull_request:44086", + "pull_request:44018|pull_request:44085", + "pull_request: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:43997|pull_request:44073", + "pull_request:44071|pull_request:44086", + "pull_request:44073|pull_request:44101", + "pull_request:44017|pull_request:44071", + "pull_request:44059|pull_request:44071", + "pull_request:44056|pull_request:44101", + "pull_request:44073|pull_request:44161", + "pull_request:44059|pull_request:44068", + "pull_request:44018|pull_request:44059", + "pull_request:44101|pull_request:44161", + "pull_request:43997|pull_request:44017", + "pull_request:43997|pull_request:44056", + "pull_request:44071|pull_request:44076", + "pull_request:44074|pull_request:44101", + "pull_request:44056|pull_request:44161", + "pull_request:43997|pull_request:44018", + "pull_request:43997|pull_request:44161", + "pull_request:44074|pull_request:44161", + "pull_request:43997|pull_request:44071", + "pull_request:44010|pull_request:44071", + "pull_request:44010|pull_request:44073", + "pull_request:44018|pull_request:44129", + "pull_request:43997|pull_request:44074", + "pull_request:43997|pull_request:44129", + "pull_request:44071|pull_request:44129", + "pull_request:44010|pull_request:44018", + "pull_request:43996|pull_request:44161", + "pull_request:43996|pull_request:44017", + "pull_request:44010|pull_request:44044", + "pull_request:44044|pull_request:44074", + "pull_request:44066|pull_request:44722", + "pull_request:44085|pull_request:44722", + "pull_request:44085|pull_request:44129", + "pull_request:43996|pull_request:44007", + "pull_request:43996|pull_request:44129", + "pull_request:44044|pull_request:44129", + "pull_request:44044|pull_request:44154", + "pull_request:44024|pull_request:44076" + ], + "prepared_review_unit_hash": "6e73c3f14aebf60382e5b31b10915d870b5e7187e677ad82c26dd1da0bd5918b", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 + }, + "budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 72, + "serialized_chars": 76942, + "estimated_input_tokens": 19236, + "estimated_eval_tokens": 38728 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "This cluster is mostly a batch of model-specific output-tracing refactors, not one single duplicate set. The true duplicate subclusters are GPT-Neo (44018/44068), GPT-J/CodeGen (44066/44085/44722), and the Flaubert-only portion of 44101/44116; most other pairs are just the same refactor theme across different models.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "44066 is the clearest anchor for the main duplicate subcluster: it directly targets #43979, touches the same GPT-J/CodeGen files as 44722/44085, and its diff matches the standardized output-tracing refactor closely.", + "best_issue_reason": null, + "best_pr_reason": "44066 is the best representative PR in the cluster because it has direct issue linkage, exact file overlap with the closest duplicates, and the most straightforward metadata for the GPT-J/CodeGen tracing change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43996", + "right": "pull_request:44085", + "accept": false, + "reason": "Different models and code paths; only the generic output-tracing theme overlaps." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44044", + "accept": false, + "reason": "CVT/FNet vs DeBERTa are unrelated concrete changes despite similar refactor style." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44085", + "accept": true, + "reason": "Same GPT-J output-tracing refactor in the same file, with the same issue target and nearly identical changes." + }, + { + "left": "pull_request:44007", + "right": "pull_request:44072", + "accept": false, + "reason": "ResNet/RegNet image backbones are separate code paths, so this is only thematic similarity." + }, + { + "left": "pull_request:44072", + "right": "pull_request:44722", + "accept": false, + "reason": "EfficientNet and GPT-J are unrelated models; no shared concrete bug/change." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44072", + "accept": false, + "reason": "CodeGen/GPT-J tracing vs EfficientNet tracing are different model implementations." + }, + { + "left": "pull_request:44013", + "right": "pull_request:44044", + "accept": false, + "reason": "MobileNetV2 and DeBERTaV2 are distinct changes with no shared code path." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J/CodeGen tracing and MGP-STR tracing are separate model-specific refactors." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-J/CodeGen and MPT touch different model stacks and different output plumbing." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44068", + "accept": true, + "reason": "Both are GPT-Neo output-tracing migrations in the same file and appear to be the same concrete refactor." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT-J/CodeGen and GPT-Neo are different model files and different changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44066", + "accept": false, + "reason": "GPT-Neo vs GPT-J/CodeGen are separate model implementations." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44722", + "accept": false, + "reason": "ResNet and GPT-J are unrelated model changes." + }, + { + "left": "pull_request:44086", + "right": "pull_request:44722", + "accept": false, + "reason": "MGP-STR and GPT-J/CodeGen are different concrete code paths." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44085", + "accept": false, + "reason": "ResNet and GPT-J are unrelated despite the shared output-tracing theme." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44019", + "accept": false, + "reason": "GPT-Neo and ResNet are separate model-specific refactors." + }, + { + "left": "pull_request:44019", + "right": "pull_request:44071", + "accept": false, + "reason": "ResNet and MPT do not share the same underlying bug/change." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT-Neo and GPT-J/CodeGen are different model families and different PRs." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated code paths." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44086", + "accept": false, + "reason": "MPT and MGP-STR are different model implementations." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44116", + "accept": false, + "reason": "GPT-Neo and Flaubert are unrelated model refactors." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are separate concrete changes." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44085", + "accept": false, + "reason": "GPT-Neo and GPT-J are different model code paths." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-Neo and MGP-STR are unrelated." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44085", + "accept": false, + "reason": "MPT and GPT-J are different model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT2 and MGP-STR are separate model-specific changes." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44086", + "accept": false, + "reason": "GPT-J and MGP-STR are different code paths." + }, + { + "left": "pull_request:44098", + "right": "pull_request:44101", + "accept": false, + "reason": "ViLT and Flaubert/XLM are different model implementations." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44722", + "accept": false, + "reason": "GPT2 and GPT-J/CodeGen are not the same underlying change." + }, + { + "left": "pull_request:44068", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT-Neo and MPT are different models and different output handling." + }, + { + "left": "pull_request:44073", + "right": "pull_request:44161", + "accept": false, + "reason": "VisualBert and LongT5 are unrelated concrete changes." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44068", + "accept": false, + "reason": "GPT2 and GPT-Neo are separate model families." + }, + { + "left": "pull_request:44059", + "right": "pull_request:44071", + "accept": false, + "reason": "GPT2 and MPT are different model code paths." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44161", + "accept": false, + "reason": "Flaubert/XLM and LongT5 are unrelated changes." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44017", + "accept": false, + "reason": "RegNet and Segformer are different backbone implementations." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44161", + "accept": false, + "reason": "RegNet and LongT5 do not share the same bug/change." + }, + { + "left": "pull_request:44074", + "right": "pull_request:44101", + "accept": false, + "reason": "TextNet and Flaubert are separate model refactors." + }, + { + "left": "pull_request:44071", + "right": "pull_request:44129", + "accept": false, + "reason": "MPT and SpeechT5 are different model families." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44071", + "accept": false, + "reason": "SqueezeBERT and MPT are unrelated concrete changes." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44073", + "accept": false, + "reason": "SqueezeBERT and VisualBert are different code paths." + }, + { + "left": "pull_request:44018", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-Neo and SpeechT5 are unrelated model changes." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44161", + "accept": false, + "reason": "CVT/FNet and LongT5 are unrelated models." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44017", + "accept": false, + "reason": "CVT/FNet and Segformer are separate implementations." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44044", + "accept": false, + "reason": "SqueezeBERT and DeBERTaV2 are different model families." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44074", + "accept": false, + "reason": "DeBERTaV2 and TextNet are unrelated changes." + }, + { + "left": "pull_request:44066", + "right": "pull_request:44722", + "accept": true, + "reason": "Same GPT-J/CodeGen standardized output-tracing change with the same files and issue target." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44722", + "accept": true, + "reason": "Both are the same GPT-J output-tracing refactor in the same file; one is just a narrower variant." + }, + { + "left": "pull_request:44085", + "right": "pull_request:44129", + "accept": false, + "reason": "GPT-J and SpeechT5 are unrelated model-specific refactors." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44007", + "accept": false, + "reason": "Different image-model code paths; only the general refactor pattern is shared." + }, + { + "left": "pull_request:43996", + "right": "pull_request:44129", + "accept": false, + "reason": "CVT/FNet and SpeechT5 are unrelated concrete changes." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44129", + "accept": false, + "reason": "DeBERTaV2 and SpeechT5 do not share the same underlying bug/change." + }, + { + "left": "pull_request:44044", + "right": "pull_request:44154", + "accept": false, + "reason": "DeBERTaV2 and ViTS are different model code paths." + }, + { + "left": "pull_request:44024", + "right": "pull_request:44076", + "accept": false, + "reason": "FocalNet and ImageGPT are different models; this is only a shared refactor theme." + }, + { + "left": "pull_request:44101", + "right": "pull_request:44116", + "accept": true, + "reason": "Both touch the same Flaubert attention/output-tracing code path; 44101 just bundles the Flaubert change with XLM." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet and stays conservative about the cluster\u2019s structure. The accepted duplicate subclusters are supported by the diff previews and shared-file/issue-target evidence, and the rejected pairs mostly stay at the level of thematic similarity across different model code paths. I don\u2019t see an overstatement that would warrant rejection." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 14, + "component_count": 27, + "review_unit_index": 2, + "review_unit_count": 2, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-43996-28", + "nodes": [ + "pull_request:43997", + "pull_request:44010", + "pull_request:44056", + "pull_request:44073", + "pull_request:44074", + "pull_request:44129" + ], + "soft_pairs": [ + "pull_request:44073|pull_request:44074", + "pull_request:44056|pull_request:44073", + "pull_request:44056|pull_request:44074", + "pull_request:43997|pull_request:44010", + "pull_request:44010|pull_request:44129" + ], + "prepared_review_unit_hash": "d07bc2bd0ba655e069bdb1888c2bb61199d8e70a2ab1f16f62e0ae67904e5160", + "component_budget": { + "node_count": 28, + "item_count": 28, + "soft_pair_count": 77, + "serialized_chars": 84510, + "estimated_input_tokens": 21128, + "estimated_eval_tokens": 42512 + }, + "budget": { + "node_count": 6, + "item_count": 6, + "soft_pair_count": 5, + "serialized_chars": 13068, + "estimated_input_tokens": 3267, + "estimated_eval_tokens": 6790 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": true, + "analyst_result": { + "summary": "These PRs are all model-specific refactors to standardized output tracing/capture, but they target different model code paths and should not be merged as duplicates.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #44129 is the strongest representative because it has an explicit linked issue and the largest, most concrete output-tracing refactor in a single model implementation.", + "best_issue_reason": null, + "best_pr_reason": "PR #44129 is the best single representative of the cluster, since it is the only one with an explicit issue target and a clearly scoped model-specific implementation change.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44073", + "right": "pull_request:44074", + "accept": false, + "reason": "Both are standardized output-capture refactors, but they touch different models (VisualBert vs TextNet) and different forward/output code paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44073", + "accept": false, + "reason": "Same refactor theme, but MPNet and VisualBert are unrelated model implementations with distinct output-tracing paths." + }, + { + "left": "pull_request:44056", + "right": "pull_request:44074", + "accept": false, + "reason": "These are separate model migrations in MPNet and TextNet; they do not fix the same concrete bug or share a mergeable code path." + }, + { + "left": "pull_request:43997", + "right": "pull_request:44010", + "accept": false, + "reason": "RegNet and SqueezeBert are different architectures and the PRs change different model-specific forward methods, so they are not duplicates." + }, + { + "left": "pull_request:44010", + "right": "pull_request:44129", + "accept": false, + "reason": "Both are output-tracing refactors, but they affect different models and different internals; this is broader stylistic similarity, not the same concrete change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "b7400b4d0eed5199b1e9dfda784df12fc4cc1e189f68fcfba4e00c7296b6e36d", + "component_budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 45, + "serialized_chars": 43791, + "estimated_input_tokens": 10948, + "estimated_eval_tokens": 22152 + }, + "budget": { + "node_count": 14, + "item_count": 14, + "soft_pair_count": 45, + "serialized_chars": 43791, + "estimated_input_tokens": 10948, + "estimated_eval_tokens": 22152 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These items are a set of model-specific output-tracing refactor PRs plus one unrelated docs toctree fix. They share a theme, but not the same concrete bug or mergeable change, so they should not be deduplicated together.", + "confidence": 0.95, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single duplicate PR stands out. If one PR must represent the cluster, 44003 is the strongest exemplar because it makes the most substantial output-tracing plumbing changes and spans two related implementations (mamba and falcon_mamba).", + "best_issue_reason": null, + "best_pr_reason": "44003 is the best representative PR for the cluster\u2019s theme: it is broader than the others, touches the core Mamba path, and includes the capture/output-recorder refactor machinery. It is still not a duplicate of the other PRs.", + "soft_edge_verdicts": [ + { + "left": "pull_request:43998", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model code paths (timm_backbone vs upernet). Same refactor pattern, but not the same underlying change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44000", + "accept": false, + "reason": "Different model families and files; both are output-tracing refactors, not one mergeable bug fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44001", + "accept": false, + "reason": "Unrelated model implementations. Similar helper usage does not make this the same fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44003", + "accept": false, + "reason": "Different architectures and code paths; this is a shared refactor idea, not the same concrete change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44029", + "accept": false, + "reason": "Separate model-specific refactors in different code paths; not a duplicate fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44001", + "accept": false, + "reason": "MobileNetV1 and UnivNet are unrelated model implementations; same refactor style only." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44002", + "accept": false, + "reason": "Different models and output types; not the same underlying bug or change." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44025", + "accept": false, + "reason": "Both refactor tracing, but for different model families and files; not mergeable as one bug fix." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44026", + "accept": false, + "reason": "Different code paths in different model wrappers; similar plumbing, not duplicate content." + }, + { + "left": "pull_request:43999", + "right": "pull_request:44027", + "accept": false, + "reason": "MobileNetV1 and SpeechEncoderDecoder are unrelated; shared tracing refactor theme is too broad." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44001", + "accept": false, + "reason": "Vision-text dual encoder vs UnivNet are different model classes; not the same fix." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44002", + "accept": false, + "reason": "Different model implementations and semantics; only the refactor style matches." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44003", + "accept": false, + "reason": "Separate model-specific output-tracing changes; not one concrete code-path issue." + }, + { + "left": "pull_request:44000", + "right": "pull_request:44025", + "accept": false, + "reason": "Different model families and output contracts; not a duplicate." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44002", + "accept": false, + "reason": "UnivNet and UperNet are unrelated; this is not the same underlying bug/change." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model code paths and outputs; too broad to treat as one duplicate fix." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44025", + "accept": false, + "reason": "DepthAnything and UnivNet are separate model refactors, not a single mergeable PR." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "UnivNet vs CodeGen are different architectures; shared output-tracing cleanup is not enough." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44029", + "accept": false, + "reason": "UperNet and RWKV are unrelated model implementations; not the same underlying issue." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44003", + "accept": false, + "reason": "Different model families and files; similar refactor pattern only." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44028", + "accept": false, + "reason": "Mamba/FalconMamba vs SuperPoint are unrelated code paths; not a duplicate change." + }, + { + "left": "pull_request:44003", + "right": "pull_request:44004", + "accept": false, + "reason": "Mamba and CodeGen are different model implementations; same broad goal, not same fix." + }, + { + "left": "pull_request:43998", + "right": "pull_request:44004", + "accept": false, + "reason": "TimmBackbone and CodeGen are unrelated; they only share a refactor motif." + }, + { + "left": "pull_request:44001", + "right": "pull_request:44004", + "accept": false, + "reason": "Different model code paths and output semantics; not mergeable as one PR." + }, + { + "left": "pull_request:44002", + "right": "pull_request:44004", + "accept": false, + "reason": "UperNet and CodeGen are unrelated model changes, not the same bug." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44026", + "accept": false, + "reason": "DepthAnything vs VisionEncoderDecoder are different model wrappers; similar tracing refactor only." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44027", + "accept": false, + "reason": "Different model families and code paths; not the same concrete fix." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44028", + "accept": false, + "reason": "DepthAnything and SuperPoint are unrelated models; no shared bug to merge." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44029", + "accept": false, + "reason": "Different implementations and output contracts; not a duplicate change." + }, + { + "left": "pull_request:44025", + "right": "pull_request:44030", + "accept": false, + "reason": "DepthAnything and DPR are different model paths; same theme, not same fix." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44027", + "accept": false, + "reason": "VisionEncoderDecoder and SpeechEncoderDecoder are distinct wrappers; not one underlying bug." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44028", + "accept": false, + "reason": "Different model subsystems; shared tracing cleanup is too broad." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model code paths and outputs; not a duplicate PR." + }, + { + "left": "pull_request:44026", + "right": "pull_request:44030", + "accept": false, + "reason": "VisionEncoderDecoder and DPR are unrelated model implementations." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44028", + "accept": false, + "reason": "SpeechEncoderDecoder and SuperPoint are unrelated; not the same concrete change." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44029", + "accept": false, + "reason": "Different model families and files; only the high-level refactor pattern matches." + }, + { + "left": "pull_request:44027", + "right": "pull_request:44030", + "accept": false, + "reason": "SpeechEncoderDecoder vs DPR are different code paths, so not a duplicate." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44029", + "accept": false, + "reason": "SuperPoint and RWKV are unrelated implementations; not the same underlying bug." + }, + { + "left": "pull_request:44028", + "right": "pull_request:44030", + "accept": false, + "reason": "SuperPoint and DPR are different model paths; no mergeable overlap beyond refactor style." + }, + { + "left": "pull_request:44029", + "right": "pull_request:44030", + "accept": false, + "reason": "RWKV and DPR are different models; the common theme is too generic to count as a duplicate." + }, + { + "left": "pull_request:44027", + "right": "pull_request:46106", + "accept": false, + "reason": "A docs toctree fix is unrelated to the SpeechEncoderDecoder model refactor." + }, + { + "left": "pull_request:44026", + "right": "pull_request:46106", + "accept": false, + "reason": "Docs navigation change vs model code refactor; clearly different work." + }, + { + "left": "pull_request:44001", + "right": "pull_request:46106", + "accept": false, + "reason": "UnivNet output-tracing refactor and docs toctree fix are unrelated." + }, + { + "left": "pull_request:44002", + "right": "pull_request:46106", + "accept": false, + "reason": "UperNet model code and documentation tree changes are not the same change." + }, + { + "left": "pull_request:43998", + "right": "pull_request:46106", + "accept": false, + "reason": "TimmBackbone refactor and docs toctree fix are unrelated artifacts." + } + ] + }, + "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": "4abfa6434f8c980164ae676378a498513b114bdddb517f66e16f62c22896f8ea", + "component_budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10910, + "estimated_input_tokens": 2728, + "estimated_eval_tokens": 5712 + }, + "budget": { + "node_count": 3, + "item_count": 3, + "soft_pair_count": 3, + "serialized_chars": 10910, + "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 pull requests are distinct model-addition efforts. They share Transformers registration/docs boilerplate, but they target different model families and code paths, so none look like duplicates.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "No single canonical PR: PenguinVL, DeepSeek-OCR-2, and Voxtral TTS are separate model integrations with different architecture, files, and implementation scope.", + "best_issue_reason": null, + "best_pr_reason": "No PR is a good global duplicate representative here; the overlap is mostly shared auto-mapping/docs scaffolding rather than the same concrete fix or model implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44662", + "right": "pull_request:45401", + "accept": false, + "reason": "Different model families and tasks: PenguinVL is a vision-language model, while Voxtral-4B-TTS is a TTS/audio model. Shared auto-registration files are generic boilerplate, not the same underlying change." + }, + { + "left": "pull_request:44662", + "right": "pull_request:45075", + "accept": false, + "reason": "PenguinVL and DeepSeek-OCR-2 are separate model integrations with different model code, tests, and docs. The overlap is only in common Transformers plumbing, not a shared bug or feature." + }, + { + "left": "pull_request:45075", + "right": "pull_request:45401", + "accept": false, + "reason": "DeepSeek-OCR-2 and Voxtral TTS are unrelated additions for different modalities and code paths. They share repository scaffolding, but not the same concrete implementation or fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "2c24f8e5c9489ec50d0ea216751d243a05e702288a9ae919d5d3d3beff1f4297", + "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 to module/kernel fusion, but they are not duplicates: #44979 introduces a standalone Module Fusion API, while #45363 wires n-to-1 kernel fusion into `KernelConfig` and model loading. The latter builds on the same theme but targets a different change surface and code path.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "#44979 is the cleaner foundational change: it defines the core `module_fusion.py` API and tests, with a focused scope around module fusion primitives.", + "best_issue_reason": null, + "best_pr_reason": "#44979 is the best single representative of the cluster because it captures the core abstraction directly, whereas #45363 is a broader integration of a different fusion mechanism.", + "soft_edge_verdicts": [ + { + "left": "pull_request:44979", + "right": "pull_request:45363", + "accept": false, + "reason": "Reject: one PR adds a general module-fusion API, the other integrates n-to-1 kernel fusion via `KernelConfig`; they are related but not the same concrete fix/change and would not be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "83a4360990b34c18ba1b52cb88c30b0c0297f4abde4049944bcccca5f5e401f8", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4414, + "estimated_input_tokens": 1104, + "estimated_eval_tokens": 2464 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both pull requests fix the same Mistral regex AttributeError in `_patch_mistral_regex` by switching from `backend_tokenizer.pre_tokenizer` to `tokenizer.pre_tokenizer`; the second PR is the more complete version because it adds test coverage.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45317 is the better canonical PR because it contains the same code fix as 45086 plus a test file, making it the more complete and reviewable change.", + "best_issue_reason": null, + "best_pr_reason": "PR 45317 is the best representative because it addresses the same concrete bug, keeps the fix localized to the same code path, and adds regression tests.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45086", + "right": "pull_request:45317", + "accept": true, + "reason": "Same underlying bug and same fix in the same file; 45317 is essentially 45086 plus tests, so they are mergeable into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 19, + "component_count": 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": "bc67d9ee3e80991c780168237506e7233fa1a8ed0353e71f6df8bd583c8441a1", + "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 are related to NemotronH, but they are not the same change. PR #45115 is a narrow conversion/weight-key refactor, while PR #45569 is a much broader model-architecture split introducing NemotronHDense/NemotronHSparse plus docs, tests, and auto-mapping updates.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "#45569 is the best canonical PR because it is the broader, more complete NemotronH rework, adding the dense/sparse model packages and associated integration work.", + "best_issue_reason": null, + "best_pr_reason": "#45569 is the best representative of the cluster; #45115 is a smaller conversion-mapping refactor and does not cover the larger architecture split.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45115", + "right": "pull_request:45569", + "accept": false, + "reason": "They share NemotronH files, but the underlying changes differ: #45115 only adjusts conversion/mapping and naming, while #45569 introduces new dense/sparse model variants, docs, tests, and auto-model wiring. They are not the same concrete PR-sized fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "c5b9ae5bdcbd043d0d1c1ab211710bf0fca1805cb8e9088eca93dcebbc04353a", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5398, + "estimated_input_tokens": 1350, + "estimated_eval_tokens": 2956 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5398, + "estimated_input_tokens": 1350, + "estimated_eval_tokens": 2956 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These PRs are related but not duplicates: one introduces heterogeneous per-layer config support, while the other builds on that groundwork to add heterogeneous model/runtime handling. They share files and concepts, but they target different code paths and could be merged separately.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45332 is the broader umbrella change, spanning cache/modeling/runtime support in addition to the heterogeneity scaffolding, so it is the better representative of the cluster.", + "best_issue_reason": null, + "best_pr_reason": "PR 45332 is the strongest overall representative because it covers the larger end-to-end heterogeneous model support path, not just config plumbing.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45332", + "right": "pull_request:45333", + "accept": false, + "reason": "Related but distinct: 45333 adds per-layer configuration plumbing, while 45332 adds heterogeneous model/cache/runtime behavior. They share groundwork but not the same concrete change and would not be merged as one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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": "f40f6aafc379625350f465d38ba32d1581ad5377fce1c15d1057f27857b8dcab", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5513, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 5513, + "estimated_input_tokens": 1379, + "estimated_eval_tokens": 3014 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are both typing-related PRs in transformers, but they address different levels of the problem: one is a broad type-checking sweep across many files, while the other adds specific typing protocols/aliases to support particular model utility code. They are not the same underlying change.", + "confidence": 0.88, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45425 is the more focused, concrete change: it introduces shared typing aliases/protocols and touches a small, coherent set of type-related call sites.", + "best_issue_reason": null, + "best_pr_reason": "PR #45425 is the better representative of a single underlying code change; PR #45415 is a broad housekeeping sweep that spans many unrelated typing cleanups.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45415", + "right": "pull_request:45425", + "accept": false, + "reason": "Both are typing-related, but #45415 is a broad type-checking cleanup across many files while #45425 adds specific typing helpers/protocols for concrete code paths. They do not look mergeable as one PR or as the same fix." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 22, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45487-5", + "nodes": [ + "pull_request:45487", + "pull_request:46101", + "pull_request:46109", + "pull_request:46135", + "pull_request:46151" + ], + "soft_pairs": [ + "pull_request:45487|pull_request:46101", + "pull_request:45487|pull_request:46151", + "pull_request:46101|pull_request:46151", + "pull_request:46109|pull_request:46151", + "pull_request:46135|pull_request:46151" + ], + "prepared_review_unit_hash": "0e268ef41b85b2782fe02c6be03fd2d57c002a5c4ebf1f1aea64ffbf3dfe6c90", + "component_budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 12529, + "estimated_input_tokens": 3133, + "estimated_eval_tokens": 6522 + }, + "budget": { + "node_count": 5, + "item_count": 5, + "soft_pair_count": 5, + "serialized_chars": 12529, + "estimated_input_tokens": 3133, + "estimated_eval_tokens": 6522 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "This cluster is not a duplicate set: it mixes a couple of model-parallel device-placement fixes with unrelated test-only expectation updates. The closest items are thematically similar, but they target different models or test suites and should not be merged as one PR.", + "confidence": 0.91, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR 45487 is the strongest representative of the cluster because it fixes a concrete model-parallel device-placement bug in real model code and spans multiple affected model implementations plus tests.", + "best_issue_reason": null, + "best_pr_reason": "PR 45487 is the best standalone PR in the group: it addresses an actual runtime bug in model code, not just test expectations, and is broader/more substantive than the other candidates.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45487", + "right": "pull_request:46101", + "accept": false, + "reason": "Both are device-placement fixes for parallel execution, but they affect different model code paths (AltCLIP/ChineseCLIP vs Qwen2 Audio) and are not the same concrete bug." + }, + { + "left": "pull_request:45487", + "right": "pull_request:46151", + "accept": false, + "reason": "One is a model-parallel runtime fix; the other only adds XPU test expectations for token classification. Different artifacts and different underlying change." + }, + { + "left": "pull_request:46101", + "right": "pull_request:46151", + "accept": false, + "reason": "Qwen2 Audio parallel fix vs pipeline test expectation updates for token classification; same broad theme of device support, but unrelated bugs/changes." + }, + { + "left": "pull_request:46109", + "right": "pull_request:46151", + "accept": false, + "reason": "Distributed trainer test genericization is unrelated to token classification expectation updates; they do not fix the same issue or code path." + }, + { + "left": "pull_request:46135", + "right": "pull_request:46151", + "accept": false, + "reason": "Both touch XPU-related tests, but for different suites and behaviors (Kosmos2.5 integration vs token classification). Not the same underlying change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 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-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": "5aafab669c6aa22fbddf6e5292c02045fc7f4e6a3540b3b9a8cd15b765411833", + "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": "The two Mamba2 pull requests target the same cached-forward seq_len>1 bug, while the Qwen3.5/OLMo hybrid PR is a different model and code path.", + "confidence": 0.94, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #46084 is the core implementation fix for Mamba2 use_cache with seq_len>1, with code and test changes directly addressing the incorrect cached forward path.", + "best_issue_reason": null, + "best_pr_reason": "PR #46084 is the best representative of the underlying bug because it is the focused code-level fix, rather than a docs-heavy follow-up.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45513", + "right": "pull_request:46084", + "accept": false, + "reason": "Different model family and bug: Qwen3.5/OLMO hybrid linear-attention caching versus Mamba2 cached forward behavior." + }, + { + "left": "pull_request:46084", + "right": "pull_request:46098", + "accept": true, + "reason": "Same underlying Mamba2 use_cache seq_len>1 bug, same issue target, and overlapping modeling/test files; one could plausibly subsume the other in a single PR." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The summary is grounded in the packet: PR #45513 is a different model/code path (Qwen3.5/OLMO hybrid linear-attention caching), while PRs #46084 and #46098 both address the same Mamba2 use_cache seq_len>1 bug and share the same issue target/files. The soft-edge verdicts are conservative and the mergeability judgment for the two Mamba2 PRs is plausible." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "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-45548-2", + "nodes": [ + "pull_request:45548", + "pull_request:45662" + ], + "soft_pairs": [ + "pull_request:45548|pull_request:45662" + ], + "prepared_review_unit_hash": "24dd820d08e706b90c2bfffa45c564b6f2b1d0486588a40f4be3703bf223d7e8", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4508, + "estimated_input_tokens": 1127, + "estimated_eval_tokens": 2510 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are related only at a high level (expert parallelism / model loading) but they fix different bugs in different code paths, so they should not be clustered as duplicates.", + "confidence": 0.93, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45662 is the stronger canonical representative because it targets a concrete loading-time corruption bug (rank-0 broadcast overwriting experts) with a focused code-path fix across the relevant loading/integration files.", + "best_issue_reason": null, + "best_pr_reason": "PR #45662 is the better standalone representative of the cluster because it addresses a specific, reproducible EP/FSDP2 failure mode rather than a more general serialization/property cleanup.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45548", + "right": "pull_request:45662", + "accept": false, + "reason": "Both mention EP and touch modeling_utils, but #45548 fixes config serialization and an EP capability flag for accelerate/ZeRO-3 loading, while #45662 fixes expert tensors being overwritten during FSDP2 loading. Different concrete bugs and merge paths." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 25, + "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": "eb845983aa4cc969553dc242db9412983f355a29a866bd19b33a9c299a278439", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4589, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4589, + "estimated_input_tokens": 1148, + "estimated_eval_tokens": 2552 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs address the same xdist-related captured_info logging problem. PR 45645 is the more complete change because it includes the testing_utils fix plus the CI log aggregation updates, while 45639 is just the core code-side piece.", + "confidence": 0.96, + "canonical_issue_reason": "Both PRs explicitly target issue #45561, which appears to be the shared tracking issue for the xdist-safe captured_info fix.", + "canonical_pr_reason": "PR 45645 is the best canonical PR because it subsumes the core testing_utils change from 45639 and also updates the workflow to read multiple captured_info files safely in CI.", + "best_issue_reason": null, + "best_pr_reason": "PR 45645 is the strongest representative of the change set: it fixes the xdist collision in code and preserves debug-log collection in the workflow, whereas 45639 only covers the code-path portion.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45639", + "right": "pull_request:45645", + "accept": true, + "reason": "Same underlying xdist-safe captured_info bug; 45645 extends the same fix rather than addressing a different change." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 26, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "cache", + "cache_hit": true, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-45737-2", + "nodes": [ + "pull_request:45737", + "pull_request:45738" + ], + "soft_pairs": [ + "pull_request:45737|pull_request:45738" + ], + "prepared_review_unit_hash": "cc2ade73a49ec6a8883a7a24c433988645588bcd049e8aad2bce56b1b24daa1c", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4131, + "estimated_input_tokens": 1033, + "estimated_eval_tokens": 2322 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "Both PRs address the same MusicgenMelody cache initialization bug in `forward()`: replacing `EncoderDecoderCache(DynamicCache, DynamicCache)` with a single `DynamicCache`. PR #45737 is the more complete duplicate because it bundles the fix with a regression test, while #45738 is the fix-only variant.", + "confidence": 0.97, + "canonical_issue_reason": null, + "canonical_pr_reason": "PR #45737 is the best canonical PR because it contains the same code fix as #45738 plus a regression test, making it the most complete representation of the bug fix.", + "best_issue_reason": null, + "best_pr_reason": "PR #45737 is the strongest overall candidate: it fixes the concrete cache bug and adds a targeted regression test, whereas #45738 only changes the implementation.", + "soft_edge_verdicts": [ + { + "left": "pull_request:45737", + "right": "pull_request:45738", + "accept": true, + "reason": "Same underlying code-path bug in `musicgen_melody` cache setup; the diffs match on the core fix and the test-only addition in #45737 could plausibly live in the same PR as #45738." + } + ] + }, + "evaluator_result": { + "accept": true, + "feedback": "The assessment is grounded in the packet: both PRs make the same `musicgen_melody` cache initialization change, and #45737 only adds a regression test on top of that fix. The soft-edge accept is conservative enough because the diffs are mergeable into one PR for the same bugfix." + }, + "evaluator_used": true, + "retried": false, + "accepted_nontrivial_soft_edge": true, + "error_kind": null, + "error_message": null + }, + { + "label": "pull_request", + "component_index": 27, + "component_count": 27, + "review_unit_index": 1, + "review_unit_count": 1, + "status": "reviewed", + "reason": null, + "source": "llm", + "cache_hit": false, + "model": "gpt-5.4-mini?service_tier=flex", + "cluster_id": "cluster-46069-2", + "nodes": [ + "pull_request:46069", + "pull_request:46073" + ], + "soft_pairs": [ + "pull_request:46069|pull_request:46073" + ], + "prepared_review_unit_hash": "71120b3c8458907102557418114771428787fce6f13761d0077f90f720cc52e5", + "component_budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4393, + "estimated_input_tokens": 1099, + "estimated_eval_tokens": 2454 + }, + "budget": { + "node_count": 2, + "item_count": 2, + "soft_pair_count": 1, + "serialized_chars": 4393, + "estimated_input_tokens": 1099, + "estimated_eval_tokens": 2454 + }, + "overflow_policy": "truncate_then_skip", + "trimmed": false, + "aggressively_trimmed": false, + "split": false, + "analyst_result": { + "summary": "These are two unrelated model-specific PRs adding attention-dispatch support to different architectures (ShieldGemma2 vs CTRL), so they should not be clustered as duplicates.", + "confidence": 0.98, + "canonical_issue_reason": null, + "canonical_pr_reason": null, + "best_issue_reason": null, + "best_pr_reason": "If a representative PR is needed, #46069 is the more substantial one (code, tests, and docs), but it is not the same change as #46073.", + "soft_edge_verdicts": [ + { + "left": "pull_request:46069", + "right": "pull_request:46073", + "accept": false, + "reason": "Different models and different concrete implementations: ShieldGemma2 support vs CTRL sdpa dispatch. They do not fix the same code-path problem and would not plausibly merge into one PR." + } + ] + }, + "evaluator_result": null, + "evaluator_used": false, + "retried": false, + "accepted_nontrivial_soft_edge": false, + "error_kind": null, + "error_message": null + } + ] +} diff --git a/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/manifest.json b/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/manifest.json new file mode 100644 index 0000000000000000000000000000000000000000..a873cb6ffaf6f920367e859d38c891df0d2641a0 --- /dev/null +++ b/snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/manifest.json @@ -0,0 +1,14 @@ +{ + "analysis_id": "hybrid-model-20260522t060021z", + "artifacts": { + "hybrid": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "model": null, + "published_at": "2026-05-22T06:10:05Z", + "repo": "huggingface/transformers", + "schema_version": 1, + "snapshot_id": "20260522T060021Z", + "variant": "hybrid" +} diff --git a/snapshots/20260522T060021Z/manifest.json b/snapshots/20260522T060021Z/manifest.json index 274e7be22f9aaf8353e7d0073afd2f116603dd4c..79cb65b19154c6237a58302e55c5b87d71ec8482 100644 --- a/snapshots/20260522T060021Z/manifest.json +++ b/snapshots/20260522T060021Z/manifest.json @@ -32,6 +32,24 @@ "timeline_events": 147 }, "extracted_at": "2026-05-22T06:00:21Z", + "published_analysis": { + "canonical_analysis_id": "hybrid-model-20260522t060021z", + "runs": { + "hybrid-model-20260522t060021z": { + "analysis_id": "hybrid-model-20260522t060021z", + "artifacts": { + "hybrid": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.json", + "hybrid_reviews": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/analysis-report-hybrid.llm-reviews.json" + }, + "channel": "canonical", + "manifest_path": "snapshots/20260522T060021Z/analysis-runs/hybrid-model-20260522t060021z/manifest.json", + "model": null, + "published_at": "2026-05-22T06:10:05Z", + "variant": "hybrid" + } + }, + "schema_version": 1 + }, "repo": "huggingface/transformers", "snapshot_id": "20260522T060021Z", "watermark": {